Restoring Rivalry As a Central Concept in Antitrust Law Harry S

Nebraska Law Review

Volume 75 | Issue 2 Article 2

1996 Restoring Rivalry As a Central Concept in Antitrust Law Harry S. Gerla University of Dayton School of Law

Follow this and additional works at: https://digitalcommons.unl.edu/nlr

Recommended Citation Harry S. Gerla, Restoring Rivalry As a Central Concept in Antitrust Law, 75 Neb. L. Rev. (1996) Available at: https://digitalcommons.unl.edu/nlr/vol75/iss2/2

This Article is brought to you for free and open access by the Law, College of at DigitalCommons@University of Nebraska - Lincoln. It has been accepted for inclusion in Nebraska Law Review by an authorized administrator of DigitalCommons@University of Nebraska - Lincoln. Harry S. Gerla* Restoring Rivalry As a Central Concept in Antitrust Law

TABLE OF CONTENTS I. Introduction ...... 209 II. Why Competition Means Rivalry Under the Antitrust Law s ...... 211 III. Why Promoting Rivalry Also Promotes Economic Efficiency, Innovation, and World-Class Competitive Industries ...... 223 A. Rivalry and the Elimination of X-Inefficiencies ..... 223 B. How Rivalry Helps Spur Innovation and Innovative Effi ciency ...... 228 C. How Promoting Rivalry Helps Create Successful World-Class Industries ...... 233 IV. The Doctrinal Consequences of Taking Rivalry Seriously ...... 237 V. Reform and Compromise ...... 249 A. Reforms-Differentiating Between Injury to Competitors (Rivals) and Injury to Competition (Rivalry)-Adoption of a Concept of De Minimis Injury to Rivalry ...... 249 B. The Grand Compromise-Giving Economic Efficiency a Co-Equal Role ...... 251 VI. Conclusion ...... 254

I. INTRODUCTION In the fractious realm of antitrust law, one proposition commands nearly universal allegiance-that antitrust laws protect "competition."1 The various statutes which utilize the term "competition" do not define it, nor do most courts applying the federal antitrust stat-

* Professor of Law, University of Dayton School of Law, B.A. Queens College, MA. University of Florida, J.D. Ohio State University. The author would like to thank Barbara Ullman Gerla, Esq. for her helpful comments and suggestions. 1. See, e.g., Copperweld Corp. v. Independence Tube Corp., 467 U.S. 752, 768 n.14 (1984); Brunswick Corp. v. Pueblo Bowl-O-Mat, 429 U.S. 477, 488 (1977); Roy B. Taylor Sales, Inc. v. Hollymatic Corp., 28 F.3d 1379, 1382 (5th Cir. 1994), cert. denied, 115 S. Ct. 779 (1995); Los Angeles Land Co. v. Brunswick Corp., 6 F.3d

209 NEBRASKA LAW REVIEW [Vol. 75:209 utes bother to define the term. 2 Nonetheless, at least until the 1970s, a strong consensus existed among antitrust courts that competition was to be defined by its dictionary meaning of rivalry among firms for the business of consumers.3 Rivalry was, in effect, the monarch of antitrust concepts. In the 1970s, however, an opposing definition of "competition" emerged. Led by judges and commentators belonging to or influenced by the Chicago School of antitrust law and economics,4 some courts began defining competition not as rivalry, but as "an allocation of resources in which economic welfare... is maximized."5 The concept that competition equals economic efficiency rather than rivalry has grown in influence over the past two decades. Many courts have either explicitly or implicitly rejected the notion that competition is rivalry and have defined competition solely in terms of economic efficiency.6 This development is so pronounced that one can fairly conclude that rivalry has been deposed from its regal position in the realm of antitrust law.

1422, 1427 (9th Cir. 1993), cert. denied, 114 S. Ct. 1307 (1995); Abcor Corp. v. AM Intl, Inc., 916 F.2d 924, 931 (4th Cir. 1990). 2. STEPHEN F. Ross, PRINCiPLEs OF ANTITRUST LAw 124 (1992); 1 SECTION OF ANTI- TRUST LAw, Am.BAR ASS'N, ANTITRUST LAw DEVELOPMENTS 42-43 (3d ed. 1992) [hereinafter ANTrrRusT DEVELOPENTS]; Sam D. Johnson & A. Michael Ferrill, Defining Competition: Economic Analysis and Antitrust Decision Making, 36 BAYLOR L. REV. 583, 585-87 (1984). 3. See, e.g., United States v. Union Pac. R.R. Co., 226 U.S. 61, 89 (1912); United States v. Aluminum Co. of Am., 148 F.2d 416, 427 (2d Cir. 1945); Commonwealth Edison Co. v. Allis Chalmers Mfg. Co., 40 F.R.D. 96, 102 (E.D. Ill. 1966); United States v. Koppers Co., 202 F. Supp. 437, 446-47 (W.D. Pa. 1962), appeal dismissed, 371 U.S. 856 (1962); United States v. Bethlehem Steel Co., 168 F. Supp. 576, 592 (S.D.N.Y. 1958); United States v. United Shoe Mach. Corp., 110 F. Supp. 295, 341 (D. Mass. 1953), affd, 347 U.S. 521 (1954). See infra note 7. 4. The Chicago School of antitrust law and economics emphasizes that allocative efficiency (or, perhaps more accurately, avoiding artificial reductions in output) is the only legitimate goal of the antitrust laws and that neoclassical microeconomic theory (with its assumptions of rational, profit-maximizing producers and consumers, the free mobility of capital and labor, relatively free entry into and exit from markets, and the inability of producers who do not possess market power to restrict output or raise prices) is the exclusive source of knowledge on whether an act or practice can reduce output. For an overview of the Chicago School of thought, see generally HERBERT HOvENKAmp, FEDERAL ANTITRUST POLICY: THE LAw OF COMPETITION AND rrs PRACTICE 61-71 (1994); Richard A. Posner, The Chi- cago School of Antitrust Analysis, 127 U. PA. L. REv. 925 (1979). See also Walter Adams & James W. Brock, Antitrust, Ideology, and the Arabesques of Economic Theory, 66 U. COLO. L. REv. 257, 282-93 (1995) (describing Chicago School principles under the rubric of "revisionist antitrust vision"). 5. Roland Mach. Co. v. Dresser Indus., 749 F.2d 380, 395 (7th Cir. 1984). See also General Leaseways, Inc. v. National Truck Rental Leasing Ass'n, 744 F.2d 588, 596 (7th Cir. 1984)("the allocation of resources that maximizes consumer welfare"). 6. See infra notes 149-57 and accompanying text. 1996] RESTORING RIVALRY TO ANTITRUST

It is time to restore rivalry to the throne and reestablish it as a central concept in antitrust law. Defining competition in terms of rivalry is both sound law and sound economics. Rivalry as competition is sound law because principles of statutory interpretation imply that competition means rivalry when that term or concept is utilized in the antitrust statutes. Rivalry as competition is sound economics because contemporary studies indicate that promoting rivalry will increase the internal efficiency of firms, spur innovation, and help develop world- class competitive industries. These developments will, in turn, enhance (or to use Judge Posner's term "maximize") the economic efficiency which the members of the Chicago School and their fellow travelers hold so dear. Rivalry cannot, however, be restored to prominence under the same terms and conditions which prevailed before it was displaced. A restored rivalry standard must avoid the past excesses (which, in part, may have been responsible for its ouster) and must accommodate itself to new political realities. First, not every injury to rivals should automatically be equated with an injury to rivalry. Second, not every minor or de minimis injury to rivalry itself should give rise to a cause of action under the federal antitrust laws. Third, those who emphasize the centrality of rivalry in the antitrust laws cannot ignore efficiency concerns, even when such concerns do not necessarily enhance rivalry. Part II of this Article is devoted to an exploration of why, as a matter of statutory interpretation, competition under the antitrust laws should be defined as rivalry. Part III of the Article discusses why fo- cusing the antitrust laws on the rivalry standard will lead to increases in productive and innovative efficiency which are so necessary for long-term success in an increasingly globalized and technologically driven economy. Part IV of this Article contains a discussion of some of the changes in modern antitrust doctrine which will be wrought by a renewed focus on using the antitrust laws to promote rivalry. Finally, Part V of the Article will discuss the doctrinal changes and accommodations which must be made if rivalry is to remain a viable central organizing concept in antitrust law.

II. WHY COMPETITION MEANS RIVALRY UNDER THE ANTITRUST LAWS

Those who claim that competition under the antitrust laws means something other than rivalry have a heavy burden of proof to bear. First, the dictionary definitions of competition all equate competition NEBRASKA LAW REVIEW [Vol. 75:209 with rivalry. 7 However, as Professor Kaplow points out, this does not decide the issue.8 But, as he also points out, none of the dictionary definitions contemplate economic efficiency as a meaning of competition. 9 Second, competition is defined as rivalry in common usage.lO The tendency to equate rivalry with competition is so strong that even Judge Frank Easterbrook, a vigorous proponent of defining competition as efficiency, cannot avoid the tendency. In his dissenting opinion in Fishman v. Estate of Wirtz,"1 Judge Easterbrook made the point that competition is efficiency, not "moment to moment rivalry."12 Yet, in the very next sentence of his opinion Judge Easterbrook writes that "[r]eal competition is bruising rivalry."13 Such is the tenacity of competition as rivalry in common usage.14 Third, and perhaps most important, for eighty years courts either explicitly or implicitly defined competition under the antitrust laws as rivalry.35 Eighty years of consistent definitional precedent should not be overturned without a compelling reason, and no such compelling reason exists. Nothing in the legislative history of any of the federal antitrust laws compels the conclusion that competition ought to mean economic efficiency rather than rivalry. However, this is not a ringing declaration that the legislative history of the Sherman Antitrust Act (or the other federal antitrust acts which, unlike the Sherman Act, actually use the term "competition" in their texts) requires that competition be defined in terms of rivalry. No such categorical declaration can be made for two reasons. First, an examination of the legislative history of the antitrust laws, particularly the Sherman Act, reveals that the individual members of Congress who passed the statute (a) had a variety of goals which they believed the statute would promote, and (b)

7. The first definition of "competition" in Webster's Third International Dictionary is "rivalry." WEBSTER'S THIRD NEW INTERNATIONAL DICTIONARY OF THE ENGLISH LANGUAGE UNABRIDGED 464 (Philip Babcock Grove & Merriam Webster eds., 1971). The Oxford English Dictionary and the Random House Dictionary are both in accord. 3 THE OxFoRD ENGLISH DIcTIoNARY 604 (2d ed. 1989); Tm RAN- DOM HOUSE UNABRIDGED DIcTIoNARY 417 (2d ed. 1993). 8. Louis Kaplow, Antitrust, Law & Economics and the Courts, 50 LAw & CoNTEMP. PROBS. 181, 210 (1987). 9. Id. 10. Id. 11. 807 F.2d 520 (7th Cir. 1986)(Easterbrook, J., dissenting). 12. Id. at 576 (Easterbrook, J., dissenting). 13. Id. at 577 (Easterbrook, J., dissenting)(emphasis added). 14. In light of the dictionary definition of competition as rivalry and the prevalence of that definition in common usage, one can only view the claim of Robert Bork, that defining competition as economic efficiency is "consistent with everyday speech," ROBERT H. BORK, TIE ANTrrRUST PARADox 61 (1978), as either incredibly disin- genuous or intellectually dishonest. 15. See supra note 3. 1996] RESTORING RIVALRY TO ANTITRUST operated in an intellectual milieu in which the term "competition" had no completely clear and singular meaning. Second, even if we could somehow ascertain the intent of the Congress in passing the Sherman Act, it is not clear that such an intent ought to be a definitive road map on how to interpret that statute. Despite the lack of clear authority in this area, there is some evidence in the legislative history of the Sherman Act that Congress intended competition to mean rivalry. The preservation of "competition" was one of the concerns voiced by Senator Sherman and others in the course of the debates over the bills that were to become the Sherman Act.16 However, none of the speakers formally defined "competition."17 Nonetheless, the speakers appeared to be using the term "competition" to mean "'rivalry,' or the presence of multiple sellers in a market."Is This, of course, supports the notion that competition under the antitrust laws means rivalry. On the other hand, at least some of the legislators who voted for the Sherman Act believed, or at least insinuated, that they were merely codifying common law prohibitions against restraints of trade in passing the Act.19 The common law at the time permitted a number of restraints of trade which clearly destroyed rivalry. For example, price-fixing agreements among tradespeople were routinely tolerated as long as they did not involve the coercion of unwilling par-

16. E.g., 21 CONG. REc. 2460 (1890) (statement of Sen. Sherman); 21 CONG. Rtc. 5957 (1890) (statement of Rep. Stewart); 21 CONG. REC. 5957 (1890)(statement of Rep. Anderson). 17. Herbert Hovenkamp, Antitrust'sProtected Classes, 88 MicH. L. REv. 1, 23 (1989). 18. Id. For example, Senator George of Mississippi indicated that competition was the act of sellers striving to sell the same article to the same customers. 21 CONG. REc. 1767 (1890) (statement of Sen. George). This, of course, closely parallels the definition of rivalry as a quest for the business of the same customers. Senator Platt similarly seemed to believe competition was rivalry when he analogized competition to "brutal" warfare among men. 21 CONG. REc. 2729 (1890) (statement of Sen. Platt). Again, the analogy smacks of rivalry as the implicit definition of competition. 19. See generally Thomas C. Arthur, Farewell to the Sea of Doubt: Jettisoning the Constitutional Sherman Act, 74 CAL. L. REv. 263, 278-81 (1986)(describing the legislative history of the Sherman Act) and sources cited therein. The great judicial champion of the view that Congress was merely codifying the common law doctrine of restraints of trade in passing the Sherman Act was Justice Holmes. See Northern Sec. Co. v. United States, 193 U.S. 197, 404 (1904)(Holmes, J., dissenting). Holmes would not, however, have dissented from the opinion that competition equals rivalry. In Holmes' view, the Sherman Act did not protect competition. It barred restraints of trade. As Holmes stated, "[tihe [Sherman] [A]ct says nothing about competition." Id. at 403 (Holmes, J., dissenting). For an analysis of Holmes' views on this point, and on antitrust law in general, see Spen- cer Weber Waller, The Antitrust Philosophy of Justice Holmes, 18 S. ILL. U. L.J. 283 (1994). 214 NEBRASKA LAW REVIEW [Vol. 75:209 ticipants to join the agreement. 20 Indeed, in the years immediately following the passage of the Sherman Act some federal courts actually allowed price-fixing agreements to pass muster under Section One of the Sherman Act because they did not involve coercion directed at third parties.2 1 This is hardly consistent with an exclusive focus on rivalry, and therefore competition, as the essence of the Sherman Act. Moreover, some of the legislators who passed the Sherman Act did not think in terms of pure "competition," but thought and spoke in terms of "fair competition."22 The concept of "fair competition" encompassed the notion that a tradesperson was entitled to a "fair" return for his efforts, even if generating such a fair return required the abate- ment or even abolition of rivalry with other tradespeople.2 3 This conceptualization of "fair competition" is inconsistent with a straightforward equivalency between competition and rivalry because it suggests that rivalry can be suspended or annulled in order to attain "fair competition." In addition to the legislative history of the Sherman Act, the intellectual climate of the late nineteenth century and early twentieth century also tends to indicate that Congress wished to protect rivalry to the extent it sought to protect competition. "Nineteenth century economic writers in general ...concentrated heavily on horizontal rivalry when they spoke of competition."2 4 Indeed many opponents and crit-

20. Herbert Hovenkamp, Labor Conspiraciesin American Law: 1880-1930, 66 Tax. L. Rav. 919, 932-34 (1988). While a few decisions upheld the de jure legality of price-fixing arrangements, most gave them de facto legality by holding that while they could not be enforced against members of the cartel, they could not be challenged by third parties or the state. Id. at 932. Professor Arthur has argued that under American (as opposed to English) common law, price-fixing cartels were de facto illegal because their inability to be enforced in a court of law rendered price- fixing agreements ineffectual, forcing cartel members to form trusts to enforce their price understandings. Arthur, supra note 19, at 282-83. 21. Herbert Hovenkamp, The Sherman Act and the Classical Theory of Competition, 74 IOWA L. Rav. 1019, 1034-35 (1989)(citing United States v. Greenhut (In re Corning), 51 F. 205 (N.D. Pa. 1892); United States v. Greenhut (In re Terrell), 51 F. 213 (C.C.S.D.N.Y. 1892); In re Greene, 52 F. 104 (C.C.S.D. Ohio 1892)). Sec- tion One of the Sherman Act bars 'every contract, combination... or conspiracy" in restraint of trade. 15 U.S.C. § 1 (1994). 22. Rudolph J. Peritz, A Counter History of Antitrust Law, 1990 DUKE L.J. 263, 266 (citing 21 CONG. REc. 2730 (1890)(remarks of Sen. Platt)). See also 21 CONG. Eec. 2461 (1890)(remarks of Sen. Sherman); 21 CONG. IEc. 3151 (1890)(remarks of Sen. Edmunds) (both speaking in terms of 'fair competition"). 23. Peritz, supra note 22, at 266. But see Edwin S. Hughes, The Left Side of Anti- trust: What Fairness Means, 77 MARQ. L. Rav. 265, 297-99 (1994)(enunciating concept of "fair competition" with more emphasis on process and linkage between superior efforts and rewards for those efforts than a guaranteed result of "fair" profit). 24. James May, Antitrust in the Formative Era: Politicaland Economic Theory in Constitutional and Antitrust Analysis, 1880-1918, 50 OMO ST. L.J. 257, 395 (1989)(emphasis added). 1996] RESTORING RIVALRY TO ANTITRUST ics of competition in general, and the new federal antitrust law in particular, implicitly assumed that competition meant rivalry.25 Once again, however, an examination of the intellectual milieu of the late nineteenth and early twentieth centuries does not totally support the notion that competition necessarily equals rivalry. First, the concept of competition may have been inextricably bound to notions of property rights and fair competition. 26 This union suggests that competition had meaning only in the context of property rights, including the right to make contracts. 27 The right to make contracts included the right to voluntarily agree not to engage in rivalry (at least on items that were not "prime necessities").28 Additionally, the concept of fair competition encompassed a right of fair return on one's property or efforts. The right to a fair return implicitly included a right to reduce or eliminate rivalry where that state of affairs infringed upon the right to a fair return. Second, by the turn of the nineteenth century at least some economic writers were beginning to break away from a definition of competition that was rooted in the concept of rivalry. These writers began to define competition in terms of end results or prevailing conditions rather than in terms of rivalry.29 This conceptualization is, perhaps, analogous to the notion that rivalry equals economic efficiency, especially the variety of economic efficiency known as allocative efficiency. Allocative efficiency is an allocation of resources in which the market price of goods and the cost of their production, including a normal return on capital, are roughly equal. 30

25. See, e.g., RIcHARD T. ELY, MONOPOLIES AND TRUSTS 142-44 (1900)(critic of the Sherman Act claiming that competition is the opposite of agreement among producers); James Logan, Unintelligent Competition a Large Factor in Making In- dustrial Concentration a Necessity, 172 N. Am.REv. 686, 688-89 (1901)(critic of Sherman Act analogizing competition to industrial warfare); John B. Clark, The Limits of Competition, 2 PoL. Scr. Q. 45, 50 (1887)(critic of anti-monopoly laws defining competition as effort to undersell rival); ArmmR J. EDDY, TnE NEW Com. PErrON 21 (4th ed. 1915)(critic of "cutthroat competition" equating competition with rivalry). On the views of contemporary critics of the Sherman Act, see generally HANs B. THORELLI, THE FEDERAL ANTruST POLICY 122-27, 311-18 (1954); 3 JosEPH DoRF AN, THE EcoNornc Mnmw iN AsmancAN CvILIzATION 119-20, 309- 11, 315 (1949). 26. See Rudolph J. Peritz, The "Rule of Reason" in Antitrust Law: PropertyLogic in Restraintof Competition, 40 HASTnqGS L.J. 285, 303-13 (1989). See also Waller, supra note 19, at 289-91 (describing Justice Holmes' view of the common law meaning of the Sherman Act in similar terms). 27. Peritz, supra note 26, at 338. 28. Hovenkamp, supra note 21, at 1030. 29. See infra notes 35-36. 30. CHmSTOPHER PAss & BYRON LowEs, Busn-mss AND MCROECoNoMncs 14 (1994). Chicago School antitrust theorists tend to use allocative efficiency in a somewhat different sense. They tend to view allocative efficiency as a state in which society's resources are employed in the uses upon which consumers collectively place NEBRASKA LAW REVIEW [Vol. 75:209

Thus, neither the legislative history of the Sherman Act, nor the intellectual context that existed at the time of its enactment unequivo- cally support the idea that competition meant rivalry. Nonetheless, the legislative history of the Sherman Act and the milieu of the time in which it was enacted do lend fairly strong support to the equation between competition and rivalry. More important, those factors do not support the view that competition ought to be defined in terms of economic efficiency. Another possible argument for efficiency supplanting rivalry as the definition of competition is that competition is an economic concept, and therefore, if economists define competition in terms of economic efficiency, the law should follow suit.3 1 Regardless of the legal merits of this argument, it does not justify defining competition as efficiency because no consensus exists among economists that competition should be defined in terms of economic efficiency. Many economists view rivalry as synonymous with competition.32 Indeed, the earliest and most elemental economic definition of competition is rivalry.33 The widespread use among economists of rivalry as a synonym for competition led Professor Kaplow to conclude that rivalry was "the" economic definition of competition.3 4 Unfortunately, among economists the tendency to draw a sharp distinction between rivalry and competition is much stronger than Professor Kaplow admits. As early as the first half of the nineteenth century, at least some economists viewed competition not as rivalry, but as an economic climate which enabled actors to attain certain eco-

the highest value. Adams & Brock, supra note 4, at 284 (citing Bom, supra note 14, at 91). 31. Professors Areeda and Turner make a somewhat similar argument in attempting to dismiss the populist noneconomic goals evident in the legislative history of the 1950 Cellar-Kefauver Act, Act of Dec. 29, 1950, ch. 1184, 64 Stat. 1125, which amended Section Seven of the Clayton Act. Section Seven of the Clayton Act, as amended by the Cellar-Kefauver Act, forbids mergers or acquisitions "in any line of commerce or in any activity affecting commerce in any section of the country, [where] the effect of such acquisition may be substantially to lessen competition, or to tend to create a monopoly." 15 U.S.C. § 18 (1994). Areeda and Turner note that the legislation which was ultimately enacted used the term "competition" and that this term was an economic concept with an economic meaning. 4 pHn. LIP AREEDA & DONALD F. TURNER, ANTrIruST LAw %904, at 13 (1980). 32. PASS & Lowas, supra note 30, at 129; CHmuSnE AimaxR & DEAN S. Azi, Dic- TiONARY OF Busmnss AND EcoNoMIcs 91 (1984); TERRY Bunca ET AL., ComEn- TIoN IN THEORY AND PRACTICE 1 (1988). See also SHARoN M. OSTER, MODERN Cowi rnr ANALYsIs 203 (1990)(implicit equation of rivalry with competition). 33. George J. Stigler, Competition, in 1 THm NEw PALGRAVE DICnoNARY oF EcoNoM- ics 531 (John Eatwell et al., eds., 1987). 34. Kaplow, supra note 8. 1996] RESTORING RIVALRY TO ANTITRUST

nomic goals.3 5 These views can be seen as precursors to the later arguments that competition should be defined in terms of an outcome, efficiency maximization, rather than a process, rivalry. By the mid-twentieth century many other economists began to equate competition with the economic model of perfect competition and began to view the former as existing when markets behaved as if they were perfectly competitive.3 6 Again, this could be taken as an endorsement of the idea that competition is maximization of economic efficiency rather than rivalry. Indeed, by 1946, before the rise of the modern Chicago School of antitrust law and economics, University of Chicago economist Frank H. Knight explicitly denied competition was a form of rivalry. Knight wrote that: The meaning of "competition" is that [actors in the market] are numerous and act individually; "atomistic" is a better word. There is no presumption of psychological competition, emulation or rivalry as this is rather contrary to the definition of economic behavior. Market relations are impersonal 3between7 persons and goods; and persuasion or "bargaining" is also excluded. Today, as economists F.M. Scherer and David Ross (neither of whom is particularly sympathetic to the Chicago School of antitrust law and economics) point out, economists generally draw a sharp distinction between the businessperson's concept of competition, i.e., rivalry, and the economist's concept of competition, a market which conforms to the model of perfect competition. 38 However, even if one believes that many modern economists do distinguish between rivalry and competition, that belief still does not justify substituting economic efficiency for rivalry as the legal definition of competition. First, competition, as it is used in the antitrust laws, is a legal concept, not an economic concept. The concept is utilized in statutes passed by legislators who likely recognized the business and ordinary person's concept of competition as rivalry rather than an economic

35. AUGUSTIN CoRNoT, MATHE ATIcAL PRINCIPLEs OF THE THEORY OF WEALTH 58-73 (1995)(originally published in 1838); NAssAu W. SENIOR, AN OuTLnm OF THE Sci- ENCE OF PoLricAL EcONOmy 102 (1836). 36. George J. Stigler, Perfect Competition, Historically Contemplated, 65 J. POL. EcoN. 1, 10-17 (1957). A perfectly competitive market possesses these characteristics: (1) "each firm is so small relative to the market that it can exert no percep- tible influence on price" (2) "all sellers must sell identical products"; (3) "there is free mobility of all resources, including free entry and exit of firms into and out of the industry"; and (4) "all buyers and sellers in the market possess complete and perfect knowledge." JAE K. SHIM & JOEL G. SIEGEL, DICIoNARY OF EcONOincs 266 (1995). 37. Frank H. Knight, New Frontiersin Economic Thought, 36 A4m. Eco. REv. 91, 102 (1946)(emphasis added). 38. F.M. SCHERER & DAVID Ross, INmusTRIAL MARimT STRucTucJ AND EcONOMIc PERFOIMANCE 15-16 (3d ed. 1990). See also REUVEN BRENNER, RIVALRY 47-48 (1987)(noting the divergence between the businessperson's concept of competition as rivalry and the economist's concept of competition as an economic climate). NEBRASKA LAW REVIEW [Vol. 75:209

concept of competition as a mimic of the theory of perfect competition. Thus, the legislators intended competition to mean rivalry. Second, many economists do still equate competition with rivalry.39 Given this fact, the fairest statement that can be made is that no consensus exists among economists as to the exact definition of competition. 40 Such an uncertain proposition hardly justifies overturning eight decades of legal precedent by substituting economic efficiency for rivalry as the legal definition of competition. Finally, even if economists widely agreed that competition is not rivalry, that agreement would not justify using economic efficiency as the definition of competition. Economists regard competition and the various forms of economic efficiency as distinct concepts.4 1 Hence, whatever competition means in economic theory, that theory does not justify making any form of economic efficiency synonymous with the concept of competition. Perhaps the strongest attack on equating competition with rivalry is that the equation is somehow "illogical." This attack has been ex- pressly launched by Robert H. Bork. Bork argues that competition cannot be given its "natural" meaning of rivalry for two reasons. First, Bork claims that if competition were defined as rivalry, the antitrust laws would forbid all acts and practices which eliminate rivalry, a result Bork finds "unthinkable" and "economically disastrous."4 2 Bork apocalyptically predicts that if rivalry were to be equated with competition, all 'Tmn[s],... partnership[s],... corporation[s] ... [and] economic unit[s] containing more than a single person" could not survive scrutiny under the antitrust laws because the creation of such entities eliminates "some kinds of rivalry between persons."43 Second, Bork objects that defining competition as rivalry makes rivalry an end in itself which must be preserved "no matter how many and how large the benefits flowing from the elimination of rivalry."44 Bork is, in essence, arguing that rivalry is not a proper object for antitrust legislation because it is a means rather than an end. Bork's first objection has a degree of surface credibility. A closer examination of the objection reveals, however, that Bork is attacking a straw man. The argument assumes that if competition is defined as rivalry, no other values or arguments may be considered in deciding

39. See supra notes 32-34 and accompanying text. 40. Paul J. McNulty, The Economic Meaning of Competition, 82 Q.J. ECON. 639, 639 (1968). The lack of precision in the definition of competition is not a new phenomenon. In the late nineteenth century, American political economist John B. Clark complained that many writers did not define the term "competition" when they used it. Clark, supra note 25. 41. Kaplow, supra note 8. 42. Bomc, supra note 14, at 58-59. 43. Id. at 58. 44. Id. 19961 RESTORING RIVALRY TO ANTITRUST 219

the legality of an act or practice under the antitrust laws. This assumption is simply contrary to logic and fact. Nothing prevents courts from considering values other than rivalry. Indeed, some forms of economic efficiency have gained parity with, and in many instances dominance over, rivalry.45 Even nonrivalrous, noneconomic efficiency considerations have occasionally triumphed in the antitrust context, despite statements by the Supreme Court and lower courts that the antitrust laws are solely concerned with competition. 46 The argument also assumes that if competition is equated with rivalry, only the immediate impact of a practice on moment-to-moment rivalry will be considered by a court, while long term or more genera- lized impact on rivalry will be ignored. This assumption is also erro- neous. Contrary to Bork's claims, equating rivalry with competition would not lead to the illegality of all economic units of more than one person. While it is true that the formation of such units automatically decreases the rivalry among the individuals making up the unit, the ability to form such units may increase rivalry in the entire market by allowing individuals already in the market to be effective rivals, or by encouraging other individuals to enter the market and to become rivals. Bork's own example of a monomaniacal devotion to rivalry contra- dicts his apocalyptic predictions and indicates that an antitrust court can look at rivalry in a broader perspective. Bork cites Justice Clark's dissent in White Motor Co. v. United States47 as an example of a single-minded and inappropriate enshrinement of rivalry. The case in-

45. See, e.g., Northwest Wholesale Stationers, Inc. v. Pacific Stationary & Printing, 472 U.S. 284, 290 (1985)(suggesting that injury to competition can be offset by gains in efficiency); Broadcast Music, Inc. v. Columbia Broadcasting Sys., Inc., 441 U.S. 1, 20 (1979)(suggesting that increase in economic efficiency is a "procompetitive effect"); Rothery Storage & Van Co. v. Atlas Van Lines, 792 F.2d 210, 229 (D.C. Cir. 1986), cert. denied, 479 U.S. 1033 (1987)(restraint on trade justified by efficiency of eliminating free-riding); The Five Smiths v. National Football League Players Ass'n, 788 F. Supp. 1042, 1055 (D. Minn. 1992)(increase in economic efficiency is a "procompetitive effect"). 46. Compare FTC v. Indiana Fed'n of Dentists, 476 U.S. 447, 463 (1986); National Soc'y of Professional Eng'rs v. United States, 435 U.S. 679, 690-91, 694-95 (1978); COMPACT v. Metropolitan Gov't of Nashville, 594 F. Supp. 1567, 1577 (M.D. Tenn. 1984); United States v. National Ass'n of Broadcasters, 536 F. Supp. 149 (D.D.C. 1982)(all denying that non-competitive values are relevant in Rule of Reason analysis) with Continental T.V., Inc. v. G.T.E. Sylvania Inc., 433 U.S. 36, 55 n.23 (1977)(dictum)(product safety as justification in Rule of Reason analysis); United States v. Brown Univ., 5 F.3d 658 (3d Cir. 1993)(spreading educational opportunity to needy students as justification in Rule of Reason analysis); Bal- moral Cinema, Inc. v. Allied Artists Pictures Corp., 885 F.2d 313, 316-17 (6th Cir. 1989)(need to countervail powerful film distributors as justification in Rule of Reason analysis); Wilk v. American Medical Ass'n, 719 F.2d 207, 227 (7th Cir. 1983), cert. denied, 467 U.S. 1210 (1984)(quality of care of medical patients allowed to override anticompetitive effects). 47. 372 U.S. 253 (1963)(Clark, J., dissenting). NEBRASKA LAW REVIEW [Vol. 75:209 volved an antitrust attack upon vertical non-price restraints.48 In reversing the grant of summary judgement for the government, the Court refused to hold that such restraints were per se illegal.49 Jus- tice Clark dissented and argued that "[tlo admit, as does the petitioner [White Motor Company], that competition is eliminated under its contracts is, under our cases, to admit a violation of the Sherman Act. No justification, no matter how beneficial, can save it from that interdiction."50 Just four years later, the Court reversed its position and held vertical territorial restraints to be per se illegal in United States v. Arnold, Schwinn & Co.51 Schwinn itself was overruled in Continental T.V., Inc. v. GTE Sylvania Inc.52 In Sylvania, the Court held that vertical non-price restraints were to be judged under the Rule of Reason.53 It gave these restraints Rule of Reason treatment in part because it was convinced that such restraints could lead to enhanced interbrand competition/rivalry which would more than offset any injuries they might cause to intrabrand competition/rivalry.54 The Court looked at overall competition/rivalry and was not blinded by the immediate impact of the challenged restraints on one particular form of rivalry. This shows that the Court is willing to accept a more general definition of rivalry rather than the narrow and restrictive formulation criticized by Bork. Bork's second reason for challenging the equation between competition and rivalry--that rivalry is a means, not an end, and therefore is not a fit object of a statute-is equally vacuous. Bork seems to be saying that Congress could not possibly have made a means the objective of a statute. This view is only sustainable if one of the following assumptions is accepted: (a) Congress never chooses means rather than ends as objects of statutes, or (b) Rivalry is an irrational means of attaining what Congress believed the ends of the antitrust laws to be, and Congress never chooses an inappropriate means to attain a statutory objective.

48. Vertical restraints of trade are restraints of trade between "firms at different levels within the chain of distribution-between, for example, a manufacturer and a wholesaler, or a wholesaler and a retailer" and "frequently are designed to limit the conditions under which firms may resell products or the conditions under which customers may purchase products." ANTrrUST DxwLoPrMNrs, supra note 2, at 99-100 (footnote omitted). Nonprice restraints are agreements which do not establish a resale price or price ranges of products or services. Cf id.at 100 (defining vertical price restrictions). 49. White Motor Co. v. United States, 372 U.S. 253, 263-64 (1963). 50. Id. at 281 (Clark, J., dissenting), cited in BoPic, supra note 14, at 58. 51. 388 U.S. 365 (1967). 52. 433 U.S. 36 (1977). 53. Id. at 59. Under the Rule of Reason, a restraint of trade will violate Section One of the Sherman Act if its anticompetitive effects outweigh its procompetitive effects. See infra note 149. 54. Continental T.V., Inc. v. GTE Sylvania Inc., 433 U.S. 36, 55-57 (1977). 1996] RESTORING RIVALRY TO ANTITRUST

Both propositions are ludicrous. Congress routinely enacts legislation with the objective of creating a means to a greater end. One example is tax subsidies to encourage investment. From 1962 to 1986, Congress gave favorable tax treatment to entities that made certain investments in plant and equipment.5 5 Congress wished to encourage those investments. The making of investments was not, however, an end in and of itself. If all that had occurred was increased investment without further effects, Congress would not have been particularly pleased. Increased investments in plant and equipment were but a means to attaining the ends of enhanced economic growth, productivity, international economic 6 competitiveness, and increased employment.5 The proposition that rivalry is irrational and could not have been intended by Congress is equally absurd. First, as will be demonstrated in Part III of this Article, the promotion of rivalry is a quite suitable method for promoting economic growth and efficiency. Sec- ond, even if this were not true, it does not mean that Congress did not choose the promotion of rivalry as the object of the antitrust laws. When drafting legislation, Congress sometimes chooses means which do not necessarily promote the ends it has in mind. Take, for example, the portion of the 1938 Federal Wage and Hours statute that requires some workers to be paid time and a half for overtime work.57 Increased pay for overtime was not the goal of the statute. Instead, the regulations were meant to increase employment opportunities and to decrease hours of work.58 Ironically, however, by creating an incentive for overtime work on the part of workers, the act may have decreased both employment and leisure time for workers.59 The proposition that Congress would not choose an inappropriate means to an end should be especially implausible to conservative political thinkers and devotees of neoclassical economics such as Bork. For more than two hundred years one of the staples of conservative political rhetoric has been the claim that liberal social legislation will often cause effects exactly opposite from those which its proponents hope to create.6 0 Thinkers in this tradition can hardly argue that

55. Revenue Act of 1962, Pub. L. No. 87-834 § 2, 76 Stat. 960, 963-66 (1962) (repealed by Tax Reform Act of 1986, Pub. L. No. 99-514 § 211, 100 Stat. 2085, 2166-70 (1986)). 56. S. REP. No. 1881, 87th Cong., 2d Sess. (1962)(reprinted in 1962 U.S.C.C.A.N. 3304, 3313-14). 57. 29 U.S.C. § 207 (1988). 58. JULIET B. SCHOR, THE OVERWORKED MERICAN 141 (1991). 59. Id. 60. ALBERT 0. HIRSCHmAN, THE RHETORIC OF REAGTION 11-42 (1991) (Hirschman terms this the "perversity thesis."). One of the examples cited by Professor Hirschman is the claim by Milton Friedman, a Nobel prize winning neoclassical economist and one of the founders of the Chicago School of economics, that mini- NEBRASKA LAW REVIEW [Vol. 75:209

Congress would not enact a statute whose ends could not be attained by the means it chose. In summary, legislative history, economic usage, and logic do not support overturning dictionary meaning, common usage, and eight decades of court precedent, all of which define competition as rivalry. Those who urge that the sole goal of the antitrust laws is the promotion of economic efficiency would be better off if they stopped trying to define competition in terms of economic efficiency and admitted that competition should not be protected at all.61 In effect, they should admit that in their zeal to avoid protecting competitors, they wrongly embraced the nostrum that the antitrust laws protect "competition, not competitors." 62 The proponents of economic efficiency can justify their change of heart on one of two bases. They could claim that the antitrust laws were never meant to protect competition and were always meant to protect economic efficiency. The problem with this approach is that it ignores much legislative history63 and more than a century of language in judicial decisions stating that antitrust laws do indeed protect competition. 64 Alternatively, economic efficiency proponents could admit that the antitrust laws perhaps were originally meant to protect competition, but that they are analagous to constitutional provisions. As such, the antitrust laws are designed to be developed by and to evolve through judicial interpretation. Under this approach, rejecting the idea that the antitrust laws protect competition could be justified on the basis that "advances" in our knowledge of economics now indicate that promoting competition will no longer foster overall economic welfare.

mum wage laws make wages fall rather than rise. Id. at 27-28 (citing MLTON FRIEDMAN, CAPITALISM AND FREEDOM 180 (1962)). 61. A possible variant of this approach is to argue that injury to competition is not enough to establish a violation of the antitrust laws and that any successful challenge to an act or practice under the antitrust laws must include a demonstration that output or economic efficiency has been injured. A panel of the Ninth Circuit recently took this approach. In Rebel Oil Co. v. Atlantic Richfield Co., 51 F.3d 1421 (9th Cir. 1995), the court admitted that competition equaled rivalry among competitors. Id. at 1433. Then, however, the court went on to say that injury to competition alone was not enough to establish a violation of the antitrust laws. The court held that an injury to "consumer welfare," defined in terms of an injury to allocative efficiency, must also be proven. Id. 62. The phrase was introduced into antitrust cases by the Warren Court in Brown Shoe Co. v. United States, 370 U.S. 294, 344 (1962). The Warren Court's antitrust jurisprudence is held in disdain by most Chicago antitrust theorists. 63. See supra notes 16-18 and accompanying text. 64. In addition to the many early cases which stressed that the antitrust laws protect competition, see supra note 1, innumerable modern courts, including courts influenced by the Chicago School of antitrust law and economics and the Supreme Court, have embraced the notion that the antitrust laws protect competition, not competitors. Id. i996] RESTORING RIVALRY TO ANTITRUST 223

This argument may be difficult for a few microeconomically-oriented antitrust theorists to accept because they strongly advocate interpreting the antitrust laws in accordance with the bare statutes, or perhaps the original intent of the legislators who enacted the antitrust laws. 65 On the other hand, many other Chicago School or microeconomically- oriented antitrust theoreticians are more inclined to accept this argument.66 The problem, however, is that the argument's basic premise is wrong. As will be demonstrated in Part III of this Article, contemporary economic studies indicate that promoting rivalry is absolutely crucial to the attainment of economic efficiency and a nation's economic well-being.

III. WHY PROMOTING RIVALRY ALSO PROMOTES ECONOMIC EFFICIENCY, INNOVATION, AND WORLD-CLASS COMPETITIVE INDUSTRIES A. Rivalry and the Elimination of X-Inefficiencies The traditional and simplest argument for how rivalry promotes economic efficiency is that rivalry forces firms to be internally efficient.67 As Learned Hand wrote a half century ago, "immunity from

65. The idea of relying on some type of original legislative intent most strongly influ- ences Robert Bork in his ultimately unconvincing attempts to prove that the Con- gress in 1890 intended the Sherman Act solely to promote allocative efficiency. See Bork, supra note 14, at 61-66. The bare meaning advocates can trace their lineage back to Justice Holmes who insisted that the Sherman Act had nothing to do with competition, but merely codified the common law of restraints of trade into federal statutory law. See supra note 19. The strongest contemporary advocate of this view is Professor Arthur. See Arthur, supra note 19; Thomas C. Ar- thur, Workable Antitrust Law: The Statutory Approach to Antitrust Law, 62 TuL. L. REv. 1163 (1988). Professor Arthur might object to this Article's characteriza- tion of him as "microeconomically oriented." However, in his proposal on interpreting the antitrust laws, a vital distinction is made between legal restraints of trade which are ancillary to legitimate business purposes and illegal restraints, i.e., those which are naked or not reasonably necessary to further the legitimate end. Arthur, supra note 19, at 337-40. Professor Arthur proposes the use of neoclassical microeconomic theory to make this crucial distinction. Id. at 341. Thus, he can be fairly characterized as "microeconomically oriented" even though he criticizes many Chicago School theorists in his articles. 66. Cf. Ricmizw POSNER, THE FEDERAL CounTs 288 (1985)(leading Chicago School theorist suggesting that the common law nature of antitrust statutes allows judges to ignore the values that the Congress meant to further in passing the antitrust laws). See also Arthur, supra note 19, at 267-68 n.9 (suggesting that both traditionalist and Chicago Schools of antitrust analysis have generally accepted the notion that the antitrust laws are of an evolutionary constitutional nature); Hughes, supranote 23, at 275-76 (arguing that most Chicago School theorists have abandoned any claim that the efficiency-centered view of antitrust law is grounded in legislative intent). 67. In economic terms, this is known as productive efficiency- obtaining the greatest amount of output from the least amount of input. PAss & Lowss, supra note 30; WALTER ADAhis & JAMEs BROCK, THE BIGNEss ComPLx 33 (1986). 224 NEBRASKA LAW REVIEW [Vol. 75:209

competition is a narcotic, and rivalry is a stimulant, to industrial progress; that the spur of constant stress is necessary to counteract an inevitable disposition to let well enough alone."6s To translate Judge Hand's observation into modern economic terms, pressures created by rivalry force firms to eliminate internal X-inefficiencies. 6 9 X-inefficiency is a "type of inefficiency which is manifest as excess unit costs of production among firms sheltered from competitive pressure."70 The argument that the antitrust laws should protect rivalry because rivalry helps reduce X-inefficiencies rests on three assumptions: (1) X-inefficiencies exist; (2) X-inefficiencies are economically significant; and (3) Competitive pressures, in the form of rivalry, will help reduce X- inefficiencies. If any of these premises is inaccurate, then the traditional view that rivalry is a tonic which helps the economy by pressuring firms to eliminate internal productive inefficiencies cannot be sustained. Each of these premises will, therefore, be evaluated. First, x-inefficiencies do indeed exist. Numerous empirical studies indicate that firms and other organizations suffer from X-inefficiencies that are often severe.7 1 The finding that firms may be operating at less than maximum efficiency, or not minimizing their costs, runs counter to a basic operating assumption of modern neoclassical economics-firms always seek to maximize their profits by maximizing output for a given input and by minimizing costs for a given output rate.7 2 Not surprisingly, the observation that X-efficiencies exist has drawn criticism from adherents to the traditional neoclassical view. Most of these critics have attacked the existence of X-inefficiency on a purely theoretical basis and without resort to prediction and testable hypothesis. The theoretical criticisms of the existence of X-inefficiency generally run along one of the following lines:

68. United States v. Aluminum Co. of Am., 148 F.2d 416, 426 (2d Cir. 1945). The relationship between competition and internal productive efficiency was noted more than two centuries ago by Adam Smith who wrote that "[m]onopoly... is a great enemy to good management, which can never be universally established but in consequence of that free and universal competition which forces everybody to have recourse to it for the sake of self-defense." ADAM SMrIr, AN INQuiRY INTO THE NATURE AND CAUSES OF THE WEALTH OF NATIONS 147 (Edwin Cannan ed., 1937). 69. Cf. Joseph F. Brodley, The Economic Goals of Antitrust: Efficiency, Consumer Welfare, and Technological Progress, 62 N.Y.U. L. REv. 1020, 1028 n.36 (1987)(noting the linkage between Judge Hand's language and the modem concept of X-inefficiency). 70. ROGER S. FRANTz, X-EmFcimNcY: THEORY, EVmENCE AND APPLICATIONs 2-3 (1988). The opposite of X-inefficiency is X-efficiency. Id. at 2. 71. For a summary of these and other studies, see FRArz, supra note 70, at 117-61. 72. PAss & LowEs, supra note 30, at 55-56; ARTHUR A. THOMPSON, JR. & JOHN P. FoRMBy, EcONOMIcs OF THE FnRm 248 (6th ed. 1993). 1996] RESTORING RIVALRY TO ANTITRUST

(a) What appear to be X-inefficiencies are really7 3 rational profit-maximizing exercises in economic rent seeking by firms. (b) What appear to be X-inefficiencies are merely the results of choices by the employees to maximize their utility by producing leisure rather than commodities sold in the market. 74 (c) Competitive capital markets will compel managers of firms to be X-efficient in order to avoid takeovers of their firms.75 (d) What appear to be X-inefficiencies are actually maximizing behaviors7 6 best explained in terms of property rights and transaction costs. All of these criticisms are flawed because they either simply assume away the problem of X-inefficiency or are based on empirically untenable assumptions.77 Take, for example, the claim that X-inefficiencies do not exist because those producing them are merely maximizing their total mix of output of leisure for themselves and commodities for the market. Those who make such a claim are engaging in what economist Harvey Leibenstein, the developer of the concept of X-(in)efficiency, terms "bull's-eye painting."7s They are taking a set of data and simply stating that any observed results are the product of rational maximizing behavior. 79 Under this approach, firms are automatically efficient. This conclusion is true even if the work ethic and management patterns at the firm produce a waste of resources, chronic absenteeism, slacking, and shoddy products.80 The judgment that such a firm is engaging in effi-

73. E.g., Mark Crain & Asghar Zardkoohi, X-Inefficiency and NonpecuniaryRewards in a Rent Seeking Society: A Neglected Issue in Property Rights Theory of the Firm, 70 A. EcoN. REv. 784 (1980); David Schap, X-Inefficiency in a Rent Seek- ing Society: A GraphicalAnalysis, 25 Q. REv. ECON. & Bus. 19 (1985). Rent seeking basically is the attempt to garner supranormal profits by obtaining monopoly power, or convincing the government to grant it to you. Gordon Tullock, Rent Seeking, in 4 ThE NE w PALGRAVEs DICrIONARY OF ECONO~Ics 147-49 (John Eatwell et al. eds., 1987). 74. E.g., E.C. Pasour Jr., Economic Efficiency and Inefficient Economies: Another View, 4 J. PosT KEYNESiAN ECON. 454 (1982); George J. Stigler, The Xistence ofX- Efficiency, 66 A. EcoN. REv. 213 (1976). 75. Thomas J. DiLorenzo, CorporateManagement, Property Rights, and the X-istence of X-inefficiency, 48 S. EcoN. J. 116 (1981). 76. Louis DeAlessi, PropertyRights, Transaction Costs, and X-Efficiency: An Essay in Economic Theory, 73 AM. ECON. Rev. 64 (1983). 77. For a detailed and extensive rebuttal to the above criticisms, see FRANrZ, supra note 70, at 183-200; Roger S. Frantz, Ex-Ante and Ex-Post Criticisms of X-Effi- ciency Theory and Literature, in STuDiEs iN ECONOMIC RATIONALrrY 43, 45-59 (Klaus Weiermair & Mark Perlman eds., 1990). 78. Harvey Leibenstein, On Bull's-Eye Painting Economics, 4 J. POST KEYNESIAN EcON. 460, 461 (1982). 79. Id. 80. This situation prevailed at all too many American firms and production facilities, at least in the 1970s and early 1980s. See generally Yoshi Tsurumi, Explaining the Japanese Paradox,N.Y. TIms, Nov. 16, 1986, § 3, at 3 (arguing that Japa- nese management share in sacrifices for productivity but American executives do not. American workers respond to this elitism with absenteeism and shoddy workmanship.); Jeffrey L. Sheler, Why So Many Workers Lie Down on the Job, 226 NEBRASKA LAW REVIEW [Vol. 75:209 cient maximizing behavior is absurd, yet it is unavoidable because the X-inefficiencies as leisure critique assumes that firms simply pick the optimal balance between leisure and commodity production.S1 Another example of a critique based on unwarranted assumptions is the claim that X-inefficiencies will be remedied by the incentives provided to managers by the capital markets, i.e., the threat of a hos- tile takeover if efficiency and profits lag. Underlying this theory is an assumption that the capital markets are perfect or perfectly efficient.82 As a general matter, the claim that some capital markets are perfect is open to serious question.8 3 Even if we assume that some markets are perfect, we still cannot simply postulate that pressure from the capital markets will remove X-inefficiencies from firms. First, the securities of many firms are not traded on large-scale public markets, the only markets for which convincing evidence of efficiency exists.84 Second, the assumption ignores the ability of incumbent managers to frustrate the ability of outsiders to take over a firm through internal devices such as poison pills and "golden parachutes" or through legislation making takeovers much more difficult.85 The other theoretical critiques of the existence of X-inefficiencies suffer from similar flaws.86 A few critics of X-(in)efficiency theory have attempted to use empirical studies to rebut the idea that firms suffer from X-inefficiencies. 87 These few empirical critiques of X-inefficiency theory have also failed to survive scrutiny.88

U.S. NEws & WORLD REPORT, Apr. 6, 1981, at 71 (suggesting that job dissatisfaction affects millions of workers and costs employers billions of dollars each year in absenteeism, reduced output, and shoddy workmanship). 81. For a more rigorous and exhaustive destruction of the x-inefficiency as leisure production theory, see FRANrz, supranote 70, at 191-93; Frantz, supra note 77, at 51-53. 82. FAN'rz supra note 70, at 164-65; Frantz, supra note 77, at 53. 83. For a summary of the empirical studies challenging the efficient market hypothesis, see Victor L. Bernard et al., Challenges to the Efficient Market Hypothesis: Limits to the Applicability of Fraud-on-the-MarketTheory, 73 NEB. L. REv. 781, 786-92 (1994). 84. See John C. Coffee, Jr., Market Failureand the Economic Case for a Mandatory Disclosure System, 70 VA. L. REv. 717, 731 (1984). See also JAMEs D. Cox ET AL., SEcuiunxs REGULATION 40-41 (1991)(questioning the evidence of the efficient market hypothesis as applied to securities traded on the over-the-counter market or the securities of smaller issuers which are traded on securities exchanges). 85. Professor (and former SEC Commissioner) Joseph A. Grundfest concludes that the confluence of management antitakeover tactics, state antitakeover laws, and financial market developments has destroyed the unfriendly takeover wave and effectively insulated inefficient management from the discipline of the capital markets. Joseph A. Grundfest, Just Vote No: A MinimalistStrategy for Dealing With BarbariansInside the Gates, 45 STAN. L. REv. 857, 858-64 (1993). 86. Frantz, supra note 77, at 45-48, 53-57; FRANrz, supra note 70, at 187, 196-99. 87. Frantz, supra note 77, at 57-58. 88. Id. 1996] RESTORING RIVALRY TO ANTITRUST

In sum, theoretical and empirical examinations confirm what those with a background in the business world realize: firms do have X- inefficiencies.S9 As economist Shlomo Maital put it, "[r]eaders with experience in the business world may be bemused that economists even question the existence of X-inefficiency. An entire branch of psychology, organizational behavior, is built on the assumption that X- inefficiency (the gap between actual and minimal costs) is alive and well." 9 0 Next, it is necessary to turn to the issue of whether x-inefficiencies are economically significant. How serious are losses caused by X-inefficiencies? A definitive answer cannot be supplied to that question because no large-scale studies on the topic have been performed. However, based on small-scale studies, Professor Liebenstein has estimated that the X-inefficiency level for the United States economy as a whole is approximately twenty percent. 9 ' In a multi-trillion dollar economy this is a vast amount of waste that is certainly economically significant.92 Indeed, the losses from X-inefficiencies probably dwarf any losses to the American economy caused by allocative inefficiencies. 9 3 The existence of economically important X-inefficiencies does not support the restoration of rivalry to a central role in antitrust law unless rivalry can play an important role in eliminating or reducing those inefficiencies. Does rivalry play such a role? X-efficiency theory has always presumed that rivalrous competitive pressures will tend to reduce X-inefficiencies, although the pressures cannot guarantee those results.94 The link between rivalry and the reduction of X-inefficiencies is certainly intuitively credible.95 Does it exist in the real world? Unfortunately, a definitive answer cannot be given.96 A large-scale, cross-longitudinal study of the issue is impossible to undertake.97

89. SCHERR & Ross, supra note 38, at 668; SHLoMO MArrAL, MINDs, M xrs AND MoNEY 112-14 (1982). 90. MArrL, supra note 89, at 112. 91. Id. at 113 (citing Harvey Leibenstein, How Inefficiency Saps CorporateProfits, FoRTuNE, June 19, 1978, at 209.). 92. MATrAL, supra note 89, at 113. Strangely, the aggregate amount of X-inefficiency in the economy has not been challenged by microeconomically-oriented theorists. Apparently they believe in a "fight them on the beaches" strategy of attempting to prove that X-inefficiency does not exist at all. 93. Brodley, supra note 69, at 1027-28 (discussing theoretical reasons why the losses from X-inefficiencies may vastly outweigh the losses from allocative inefficiencies); HARVEY LEMENSTEIN, BEYoND EcoNoMIc MAN 249-51 (1976). 94. LEMENSTEIN, supra note 93, at 207-08. 95. See Rex J. Ahdar, Antitrust Policy in New Zealand: The Beginning of a New Era, 9 INTL. TAx & Bus. LAw. 329, 343 (1992). 96. Id. 97. F.M. Scherer, Antitrust Efficiency, and Progress, 62 N.Y.U. L. REv. 998, 1004 (1987). 228 NEBRASKA LAW REVIEW [Vol. 75:209

Nonetheless, numerous small-scale studies have been consistent in demonstrating that the presence of the pressures generated by rivalry do tend to diminish the amount of X-inefficiencies in the firms sub- jected to those pressures.98 These studies do support the intuitive view that rivalry can reduce X-inefficiencies and promote appreciable improvements in overall economic efficiency.

B. How Rivalry Helps Spur Innovation and Innovative Efficiency Intuitively, rivalry would seem to spur firms to bring forth innovations to help them stay in front in the competitive race.99 For a number of years, however, some microeconomic theorists have challenged the proposition that rivalry breeds innovation and innovative efficiency. These theorists claim, in fact, that strong rivalry actually inhibits innovation. 00 If these assertions are correct, then promoting rivalry through the antitrust laws might not be a sensible economic or industrial policy. The well-being of advanced industrial economies is, in large part, dependent upon innovation and innovative efficiency.O1 This importance is described by Professor Brodley: Innovation efficiency or technological progress is the single most important factor in the growth of real output in the United States and the rest of the industrialized world. Indeed, studies have shown that over the forty year period from the late 1920s to the late 1960s, at least half of the gain in1 0United2 States output was due solely to technological and scientific progress. Thus, a charge that rivalry inhibits innovation or innovative efficiency is a very serious accusation. An examination of the theories which suggest that rivalry inhibits innovation demonstrates, however, that rivalry in fact assists innovation more than it retards it and that attempts in the real world to spur innovation by limiting rivalry are likely to fail. Two theories on how "excessive" rivalry hinders innovation have been developed. The first theory is that rivalry (or competition) sty- mies innovation by diminishing a firm's ability to capture the rewards of its innovations.-0 3 Innovation is often an expensive process, and

98. For a summary of studies indicating a link between lack of competition and X- inefficiencies, see SCHERER & Ross, supranote 38, at 668-72; FRAz, supra note 70, at 161-81. 99. This intuition is also reflected in Learned Hand's dictum that rivalry is a stimulant to "industrial progress." See supra text accompanying note 68. 100. See infra notes 103-08, 121-24 and accompanying text. 101. Brodley, supra note 69, at 1026. 102. Id. (footnotes and citations omitted). 103. See Thomas M. Jorde & David J. Teece, Innovation, Cooperation, and Antitrust, in ANTITRUST, INNOVATION, AND COMPETITIvENESS 47, 52-54 (Thomas M. Jorde & David J. Teece eds., 1992). 1996] RESTORING RIVALRY TO ANTITRUST always an uncertain one. For each successful innovation, scores of expensive "dry holes" will have to be explored.104 In order for a private firm to engage in innovation, it must believe that it will be able to recapture the costs of innovation through enhanced future profits, or in economic terms, it must be able to capture "economic rents" from the innovation in the future.105 Unfortunately for many would-be innovators, innovations often have some of the characteristics of public goods and profits from the innovation can be siphoned off by free-riding imitators.' 06 The greater the rivalry or competition in the marketplace, the more likely that free-riding imitators will siphon off the profits. Furthermore, the presence of extensive competition or rivalry in the market makes it less likely that the innovator will be able to capture the "economic rents" arising from its innovation.10 7 Thus, fierce competition in the market supposedly decreases innovation by diminishing the incentive for private firms to engage in the process. Empirical data exists which gives some modest support to this theory.1OS A closer examination of both the theory and the empirical evidence supporting it reveals a much more complex relationship between rivalry (or competition) and incentives to innovate. On a theoretical level, a number of factors, such as first mover advantage, brand loyal- ties, competitor inertia, and risk aversion may give an innovator the ability to reap substantial profits from an innovation despite the seeming inability to capture economic rents because of fierce competition.10 9 In addition, rivalry may promote the elimination of internal

104. Thomas M. Jorde & David J. Teece, Rule of Reason Analysis of Horizontal Ar- rangements: Agreements Designed to Advance Innovation and Commercialize Technology, 61 ANTurRUST L.J. 579, 581-82 (1993). 105. "Economic rents" are returns on an asset which are greater than the minimum returns necessary to keep the asset in production. See G.L.S. SHAcKLE, ECONOM- ics FOR PLEASURE 116-17 (2d ed. 1968); ERwiN ESSER NEI~mRs, DITIONARY OF BUsmNEss AiN EcONOMIcs 142 (3d ed. 1974). 106. Jorde & Teece, supra note 104, at 594. 107. See William J. Baumol & Janusz A. Ordover, Antitrust: Source of Dynamic and Static Inefficiencies?, in ANTrIRUST, INNOVATION AND CoiETrr"vNIss 82, 83-84 (Thomas M. Jorde & David J. Teece eds., 1992). 108. See, e.g., WIuhm L. BALDwIN & JoHN T. SCOTT, MAMET STRUCTRm AND TECH- NOLOGICAL CHANGE 128-44 (1987); Edwin Mansfield, Composition ofR and D Ex- penditures: Relationship to Size of Firm, Concentrationand Innovative Output, 63 REv. ECON. & STAT. 610, 610-13 (1981); Richard C. Levin et al., R & D, Ap- propriability, Opportunity, and Market Structure: New Evidence on Some SchumpeterianHypotheses, 75 AM. ECON. REv. 20, 20-24 (1985). 109. MICHAEL E. PoRTER, TsE Coan'rmv ADVANTAGE OF NATIONS 788 n.61 (1990). First mover advantage is a "competitive advantage held by a firm by virtue of being first in a particular market or first to use a particular strategy." OSTER, supranote 32, at 364. Brand loyalty is "[a] consistent faithfulness in the choice of a specific product over comparable products." Sonw & SmGEL, supra note 36, at 39. NEBRASKA LAW REVIEW [Vol. 75:209 inefficiencies which hinder the ability of firms to profit from innovations they create. More important, the relationship between rivalry and innovation is not linear. Instead, rivalry and competition strongly enhance innovative efficiency, at least up to a certain point. At that point, increases in rivalry or competition appear to decrease innovation, perhaps because of the process described above.11o If we were omniscient beings, we might be able to discern exactly where that magical point was for any given industry and adjust competition policy to produce the ideal amount of competition or rivalry to optimize innovation. Unfortunately, with our bounded rationality and incomplete information, we cannot fine tune public policy with that degree of accuracy. Any policy we pursue is likely to produce too much or too little rivalry to maximize innovative efficiency. The empirical data indicates that it is more dangerous for innovation if we err on the side of too little rivalry than if we err on the side of too much rivalry.111 This point is best made by economists F.M. Scherer and David Ross who write "[w]hat is needed for rapid technical progress is

Competitor inertia simply refers to the tendency of organizations to "resist innovation." DON HELLRIEGEL & JOHN W. SLOCUM, JR., ORGANIZATIONAL BEHAV- IOR 552 (2d ed. 1979). These tendencies are caused by a number of factors including threats to power and influence, organizational structure, limitations on resources, and resources that are unable to be switched because they are commit- ted to existing projects and interorganizational agreements, such as labor contracts and commitments to suppliers. Id. Risk aversion is preferring "asure thing even when a risky thing is the better bet." MORTON HUNT,THE STORY OF PSYCHOLOGY 547 (1993). 110. SCHERER & Ross, supra note 38, at 645-46. The relationship between competitive pressures and innovations may also be explained by human behavioral psychology. For almost ninety years psychology has recognized what is known as the Yerkes-Dodson Law. The Yerkes-Dodson Law "postulates an upside down U- shaped relationship between the quality of decision making and the pressures on an individual/organization. Initially, the pressure improves the quality of decision making but after a point, additional pressure can lower the quality of decision making." Harinder Singh, Economic Behavior, in 2 ENCYCLOPEDIA OF HUMAN BEHAVIOR 203, 206 (V.S. Ramachandran ed., 1994). Competitive pressures may increase the ability of managers to make correct decisions on innovation but, if those pressures become too intense, the managers may start making poor decisions resulting in a decrease in innovations. For an attempt to utilize the Yerkes-Dodson Law to explain partial X-inefficiencies in firms, see HARVEY LEIBENSTEIN, INSIDE THE FIRM: THE INEFFICIENCIES OF HIERARCHY 18-20, 232 (1987). 111. P.A. Geroski, Innovation, Technological Opportunity, and Market Structure, 42 OxFoRD EcONOMIc PAPERS 586, 600 (1990). Cf Robert P. Merges & Richard R. Nelson, On the Complex Economics of Patent Scope, 90 COLUM. L. REv. 839, 908 (1990) (Despite inefficiencies, rivalry produces more technological advances than a regime in which development is controlled by one or a few sources.). 1996] RESTORING RIVALRY TO ANTITRUST 231 a subtle blend of competition and monopoly, with more emphasis in general on the former than the latter... "112 Cognitive psychology may provide an explanation of why it is better for innovation to err on the side of too much rivalry. Engaging in innovation is a risky process. 113 Psychological studies have consist- ently indicated that a person's attitude toward risk under uncertain conditions tends to be determined by how she frames her situation.114 If a decisionmaker frames her choice as an opportunity for gain, she tends to be risk averse (risk avoiding). On the other hand, if she frames her situation as one in which she is trying to avoid a certain loss, she tends to be risk affinitive (risk preferring).1 35 In the absence of competition, a decision on whether to undertake the research and development efforts necessary to create an innova-

112. SCHERER & Ross, supra note 38, at 660. The authors' conclusion is strongly but- tressed by the recent empirical work of Professors Merges and Nelson who sum- marize their data in the following statement: Public policy, including patent law, ought to encourage inventive rivalry, and not hinder it. As the "race to invent" models show, a rivalrous structure surely has its inefficiencies. But such a structure does tend to generate rapid technical progress and seems a much better social bet than a regime where only one or a few organizations control the development of any given technology. Merges & Nelson, supra note 111. In a more recent essay, Merges and Nelson specifically apply their findings to the antitrust laws. They write: [O]ne question to ask about proposals for reform of antitrust law is whether they would tend to unduly contract the number of capable and motivated rivalrous sources of invention.... [Olur general message is to be wary of arguments that say it is not important to preserve, or if need be create, real rivalry in invention and innovation. Robert P. Merges & Richard R. Nelson, Market Structure and TechnicalAdvance: The Role of Patent Scope Decisions, in ANrrusT, INNoVATIoN Am CoETInrV- NEss 185, 218 (Thomas M. Jorde & David J. Teece eds., 1992). 113. See Jorde & Teece, supra note 104. 114. Amos Tversky & Daniel Kahneman, The Framing of Decisions and the Psychol- ogy of Choice, 211 SCIENCE 453, 457-58 (1981). 115. See, e.g., PAUL J.H. ScHOEmAF R, ExPERn NTs ON DECISIONS UNDER RISK: TH EXPECTED UTrnrrY HYPOTHESIS 45-90, 109-25 (1980); John W. Payne et al., Behav- ioral Decision Research: Constructive Processing Perspective, 43 ANN. REV. PSYCHOL. 87, 95-97 (1992); Amos Tversky & Daniel Kahneman, Advances in Pros- pect Theory: Cumulative Representationof Uncertainty,5 J. RISK & UNCERTA1rr 297 (1992); Don N. MacDonald & Jerry L. Wall, An Experimental Study of the Allais Paradox Over Losses: Some PreliminaryEvidence, 28 Q.J. Bus. & ECON. 43 (1989); Dan J. Laughhunn et al., ManagerialRisk Preferences for Below-Tar- get Returns, 26 MGwrr. ScL 1238, 1247-49 (1980). But see N.S. Fagley & Paul M. Miller, The Effects of Decision Framingon Choice ofRisky vs. Certain Options, 39 ORGANIZATIONAL BEHAv. & Hmi. DECISION PROCESSES 264 (1987) (empirical results suggesting that subjects avoid risk in situations when attempting to avoid near certain loss). For a comprehensive review of studies and the impact of those studies on the discipline of economics, see generally Lola L. Lopes, Psychology and Economics: Pespectives on Risk, Cooperation,and the Marketplace, 45 ANN. REv. PSYCHOL. 197, 199-205 (1994). NEBRASKA LAW REVIEW [Vol. 75:209 tion is likely to be framed as an opportunity for gain. Therefore, the managers making the decision are more likely to be risk averse and hesitant to "take the plunge" into innovation through research and development. In contrast, in a market characterized by fierce competition and rivalry, managers may well frame the decision to commit to a research and development program as necessary to avoid a loss. They may reason that if they do not innovate, at least one of their many fierce rivals will, and their firm will suffer a loss by not innovating.1 16 Thus, the managers in an atmosphere of fierce competition and rivalry will tend to be more risk affinitive and more willing to undertake the risks of the innovative process. The other economic theory which suggests that rivalry is injurious to innovation is what Professor Kitch has termed "prospect theory."117 Those who subscribe to this theory hypothesize that innovations form a common pool, much like a fishing hole.138 A limited number of in- ventions exist. As a firm discovers an invention it decreases the total available to other firms. This encourages firms to "race" for inven- tions. This leads to "overfishing" and wasteful duplication of inventive efforts.119 Professor Kitch has argued that prospect theory suggests that rivalry in innovation is economically inefficient and that nonrivalrous control over innovation is superior.120 Kitch proposes the traditional neoclassical solution for a common pool problem, assigning specific property rights to the items in the pool.121 In effect, Kitch suggests that rivalry for innovations be curbed and that innovations be controlled by a single source, or at least a limited number of sources. Professor Kitch's thesis is flawed in several ways. First, he completely ignores the incentive effects of rivalry in pressuring firms to

116. The word "may" is used quite deliberately for two reasons. First, psychological evidence is empirical and probabilistic, not hypothetical and determinative. Hence, psychology does not deny the reality that some people who frame a situation as an opportunity for gain will be risk affinitive and some people who frame a situation as avoiding a certain loss will still be risk averse. Second, the managers may not believe that any danger exists in not innovating. They may believe that the situation is one in which it is better for rival firms to take the risk while they reap the benefits by free-riding and imitating the rivals' innovations. See SHLOMO MAITAL & SmiARoNE L. MArrAL, EcONbMc GAuMs PEoPLE PLAY 88-91 (1984)(a game theory representation of the above reasoning). On the other hand, the managers may indeed believe that if they do not pursue and take advantage of innovation, one of their rivals will. For a theoretic game representation of this reasoning, see id. at 86-88. 117. Edmund W. Kitch, The Nature and Function of the Patent System, 20 J.L. & EcoN. 265, 266 (1977). 118. Id. at 265. 119. Id. at 265-66. 120. Id. at 285-86. 121. See id. at 275. For a critique of many of the underlying assumptions of this theory and a proposed solution, see Merges & Nelson, supra note 115, at 871-75. 19961 RESTORING RIVALRY TO ANTITRUST 233

create innovations and in forcing them to eliminate the internal inefficiencies which prevent them from exploiting those innovations. 122 Second, the world of innovations which Professor Kitch describes simply does not exist in many industries. In many industries, innovations are not discrete developments whose value is captured by the first party to develop it. Instead, technical advance is what Professors Merges and Nelson call "cumulative." In these industries, "today's advances build on and interact with many other features of existing technology."123 Locking up technology in such industries by granting strong property rights will slow rather than accelerate technological advances in such industries. Third, the empirical studies of Professors Merges and Nelson, although anecdotal, give strong support to the notion that an atmosphere of rivalry is much more conducive to innovations than a regime in which one or a few sources are given complete control over the development of innovations.1 24 Professor Kitch's prospect theory does not rebut the observation that rivalry is good for innovation, let alone justify placing restrictions on rivalry in the name of spurring innovation. In sum, the intuitive belief that rivalry spurs innovation is both theoretically and empirically sound. While some economic theories seek to challenge the correlation between rivalry and innovative efficiency, those theories are either empirically flawed or useless for for- mulating a real world policy which is apt to enhance the flow of innovations. In terms of innovative efficiency, promoting strong rivalry is sound economic policy.

C. How Promoting Rivalry Helps Create Successful World- Class Industries Promoting rivalry is also sound economic policy because vigorous domestic rivalry fuels the growth of world-class industries which are

122. See supra notes 94-98 and accompanying text. 123. Merges & Nelson, supra note 111, at 881 (footnote omitted). Of course, observers can certainly differ as to whether an industry is indeed characterized by discrete technologies whose economic rents must be appropriable to spur innovation or by cumulative technologies where innovations must not be locked up and incentives to rivalry must be maintained. Compare Anthony L. Clapes, Confessions of an Amicus Curiae: Technophobia, Law, and Creativity in the Digital Arts, 19 U. DAYTON L. Rlv. 903, 949-50, 962 (1994)(emphasizing the need to maintain the lure of profits as an incentive to innovate in high technology industries in general and the computer software industry in particular) with Marshall Leaffer, Engi- neering Competitive Policy and Copyright Misuse, 19 U. DAYTON L. REV. 1087, 1094-95 (1994)(arguing that the computer software industry is characterized by cumulative technologies which must be open for exploitation if innovation is to continue). 124. Merges & Nelson, supra note 111, at 877. NEBRASKA LAW REVIEW [Vol. 75:209

able to compete and dominate in the global marketplace. The empirical studies of Professor Michael Porter of Harvard University's School of Business Administration strongly support this contention. In studying the industries in ten nations, Professor Porter found that firms from particular nations tended to dominate industries or industry segments.1 25 He identified four interrelated attributes or determinants which were key to establishing the dominance of various nations in different industries. The four attributes Professor Porter identified are: 1. Factorconditions. The nation's position in factors of production such as skilled labor or infrastructure necessary to compete in a given industry. 2. Demand conditions. The nature of home demand for the industry's product or service. 3. Related and supportingindustries. The presence or absence in the nation of supplier industries and related industries that are internationally competitive. 4. Firm strategy, structure, and rivalry. The conditions in the nation gov- erning how companies are created, organized, and managed, and the nature of domestic rivalry.1 2 6 While all four determinants work together in creating a competitive advantage for particular national industries, the presence of strong domestic rivalry is the "first among equals" in Professor Porter's hierarchy.127 Vigorous domestic rivalry is not only important as one of the determinants of national competitive advantage, but also as a catalyst for the development of the other determinants.128 Professor Porter's research lends powerful support to the idea that antitrust laws ought to be used to promote rivalry. The promotion of domestic rivalry is vital to the competitive success of a nation's industries, and therefore to the economic well-being of that nation. Thus, promoting domestic rivalry is sound economics. As Professor Porter specifically recognizes, forceful enforcement of strong antitrust laws is perhaps the most important tool governments have for facilitating the creation of strong domestic rivalry.129 Professor Porter's view has not, however, gone unchallenged. Pro- fessor David J. Teece of the Haas School of Business at the University of California at Berkley maintains that Professor Porter misreads his own anecdotal data and overemphasizes the role of rivalry in spurring dominant national industries.130 Professor Teece argues that cooper-

125. The nations Professor Porter studied were the United States, Japan, Germany, Denmark, Italy, Korea, Singapore, Sweden, Switzerland, and the United King- dom. PORTER, supra note 109, at 21. 126. Id. at 71 (footnote omitted). 127. ToM PETERs, LIBERATION MANAGEMENT 504 (1992). 128. See Porter, supra note 109, at 560. 129. Id. at 662-64. 130. David J. Teece, Information Sharing, Innovation, and Antitrust, 62 AN'rrrusT L.J. 465, 471 n.21 (1994). 1996] RESTORING RIVALRY TO ANTITRUST 235

ation among competitors is as important as rivalry among them to produce innovation and economic progress.1S1 Professor Teece's criticism of Professor Porter's conclusions is somewhat misplaced. Professor Teece recognizes that Professor Porter does not endorse a world of unthinking and unremitting rivalry in which every competitor refrains from doing anything which might help a rival firm.' 3 2 Professor Porter, like Professor Teece, recognizes that some cooperation among competitors is needed to build dominant globally competitive industries. For example, Professor Teece emphasizes the need for cooperation among competitors in the area of information exchange for purposes such as standard setting and benchmarking.133 Professor Porter likewise believes that information exchanges are vital in building dominant competitively successful industries. He emphasizes that "the exchange and flow of information about needs, techniques, and technology among buyers, suppliers, and related industries" is necessary for the functioning of the determinants of competitive advantage.134 Thus, Professor Teece's quarrel with Porter is not based on the idea that Porter totally ignores the need for some cooperation among competitors. What underlies Professor Teece's criticism of Professor Porter is that Professor Teece has a fundamentally different view of the conditions which are necessary to spur innovation and industrial advancement. Professor Teece believes that in order to spur innovation and con- comitant industrial advancement, the incentive for firms to innovate and commercially exploit those innovations must be preserved to the maximum extent possible. In Professor Teece's view, one of the greatest threats to the incentive to innovate is the inability to capture rents from innovation because of imitation by free-riding rivals.135 While Teece acknowledges the need for rivalry,13 6 he is concerned about its potential for blunting incentives to innovate. Thus, for example, Teece (and his collaborator, Professor Thomas M. Jorde) is willing to tolerate agreements among competitors which "secure appropriability and prevent free-riding and opportunistic behavior" with respect to innovations which cannot otherwise be protected under existing intellectual property law.'37

131. Id. Professor Teece does not, however, deny the point that rivalry is important to innovation. He merely suggests that a blend of rivalry and cooperation is needed to maximize innovation. Id. 132. Id. at 471. 133. Id. at 469-81. 134. PORTR, supra note 109, at 152-53. Professor Teece cites this passage in his recent article. Teece, supra note 130, at 471. 135. Jorde & Teece, supra note 103. 136. Teece, supra note 130, at 471 n.21, 472. 137. Jorde & Teece, supra note 104, at 617. NEBRASKA LAW REVIEW [Vol. 75:209

Professor Porter does not deny that incentives to innovate are important.'38 However, he suggests several reasons why strong rivalry is not inconsistent with innovation. First, in imperfect markets, innovators can gain a significant competitive advantage in spite of firms which imitate their innovations.3s 9 Second, he notes that rivalry cre- ates a penalty for not innovating if others do so. Professor Porter's research indicates that this penalty is at least as important as positive incentives to innovate "because of the inertia and complacency of many companies."140 Third, Professor Porter's theory suggests that rivalry, because of its beneficial impact on the determinants of national competitive advantage, will put firms in a better position to exploit successfully any innovations they create.'41 Research by a number of scholars indicates that bureaucratic sclerosis and other internal firm inefficiencies are frequently responsible for a failure to create innovations or profitably exploit those innovations which are created.' 42 To the extent that rivalry curbs X-inefficiencies in firms,143 it can serve to facilitate the creation and commercial exploitation of innovations. Professor Porter's views on innovation differ in an even more fundamental way from Professor Teece's views. He does believe that some incentives to innovate must be preserved in order to spur the creation of innovations and technological advances. However, Porter does not believe that the creator of an innovation must be able to wring every last penny of profit from an innovation or completely block competitors from obtaining any uncompensated benefits from the innovation. First, such arrangements are unnecessary to create incentives to innovate in light of the competitive advantages which can stem from innovations and the threat of innovations by active rivals. Second, and even more fundamental, a system which completely blocks competitors from obtaining unpaid-for benefits from a rival's innovation or advance would, in the long run, reduce the total number of innovations and technological advances. The total number would be reduced because innovation is a cumulative process which is enhanced by the widespread diffusion of knowledge among competitors and their ability to build upon and utilize earlier innovations created by their rivals. A system which allows competitors to erect insur- mountable barriers to the utilization of their advances by rivals would

138. PORTER, supra note 109. 139. Id. 140. Id. The psychology of risk-taking may explain the phenomenon observed by Pro- fessor Porter. See supra notes 114-15 and accompanying text. 141. PORTER, supra note 109. 142. Geroski, supra note 111, at 587-88; SCHERER & Ross, supra note 38, at 652-54; MORTON I. KAmIEN & NANCY L. SCHWARTZ, MARKET STRuCrRE AND INNOVATION 67-68 (1982). 143. See supra notes 94-98 and accompanying text. 19961 RESTORING RIVALRY TO ANTITRUST

hinder the diffusion of knowledge and experimentation so necessary to the innovative process.144 When viewed in this light, Professor Teece's critique of Professor Porter's theory is nothing more than a rehash of the traditional argument that rivalry is destructive of incentives to innovate.145 Professor Teece's critique suffers from all the weaknesses of that line of argument; the critique (a) ignores the valuable incentive effects of rivalry, (b) cannot be the basis of a workable public policy, and (c) assumes a world of discrete innovations which does not correspond to the cumulative nature of innovation in the real world.146 Moreover, Professor Teece's critique is also inconsistent with his own recognition of the importance of the free flow of information for innovation and the creation of world-class competitive industries.147 The view that firms must be able to capture the lion's share of rewards from their advances and innovations is incompatible with the observation that industries, in order to be globally dominant, must be able to absorb and utilize information about techniques and technology developed by their competitors. Professor Teece's criticism of Professor Porter's thesis is, in the end, unconvincing. Professor Porter's fundamental point stands. Strong domestic rivalry is vital to the creation of globally competitive industries. The use of the antitrust laws to promote rivalry is, therefore, sound economic and industrial policy.

IV. THE DOCTRINAL CONSEQUENCES OF TAKING RIVALRY SERIOUSLY The restoration of rivalry to a central position in antitrust law will have a strong impact on contemporary antitrust doctrine in a number of areas. The restoration will be felt most profoundly in the area of Section One of the Sherman Act.

144. Cf PoRTER, supranote 109, at 635 (noting that while some protection for intellectual property is a necessity, overly strong intellectual property protection locks firms into yesterday's technologies and hinders the diffusion of know-how necessary for further innovations). In contrast, Professors Teece and Jorde's objection to intellectual property protections for innovations is not that they hinder the diffusion of knowledge, but that they are basically ineffectual in curbing free- riding. Jorde & Teece, supra note 104, at 583. 145. See supra notes 94-98, 103-08 and accompanying text. 146. See supra notes 103-16, 122-24 and accompanying text. In the high tech industries which are of particular concern to Teece, loss of innovation through "excess rivalry" is particularly unlikely. In such industries, technological opportunities are rich because the relevant science and knowledge bases move rapidly and often unpredictably. In industries with these characteristics, empirical studies tend to demonstrate that the appropriability of quasi-rents from innovation is not diminished by "excessive rivalry." SCHERR & Ross, supra note 38, at 647. 147. See Teece, supra note 130, at 471. 238 NEBRASKA LAW REVIEW [Vol. 75:209

The primary tool for analyzing whether a restraint passes muster under Section One of the Sherman Act is the Rule of Reason. Under the Rule of Reason, a restraint of trade will violate Section One if its anticompetitive effects outweigh its procompetitive effects.148 If, as this Article argues, rivalry is synonymous with competition, then, by definition, a restraint which diminishes rivalry has anticompetitive effects. The preceding point may appear to be obvious or even simplistic. Yet, it calls into question a number of contemporary trends in the interpretation of Section One of the Sherman Act. The doctrine which would be most affected by a renewed emphasis on rivalry is the rule articulated by many courts that in order to prevail on a Rule of Reason the plaintiff must show that the practice in question has an claim, 149 adverse impact on the price or total output of goods or services. This rule is based on the notion that only the various forms of economic efficiency (or even more narrowly, allocative efficiency or output itself) matter in antitrust law. This notion is, of course, totally inconsistent with an emphasis on competition defined as rivalry. Requiring proof of an increase in price or a decrease in output as a sine qua non for an injury to competition is justifiable only if the vast majority of significant decreases in marketplace rivalry result in a dis- cernable increase in price or decrease in output. A diminution in rivalry, however, is often reflected not so much in an immediate increase in price or decrease in output, but in (a) an increase in internal inefficiencies (X-inefficiencies) within firms,15o (b) a decrease in innovations by firms,1 5 1 or (c) an increase in managers' arrogance toward customers and fellow employees.152 In the long term, these phenomena will likely lead to increases in price (or price equivalent) and decreases in output. These long-term effects may not manifest themselves for an extended period of time. Requiring plaintiffs to establish an increase in price or decrease in output ignores this long- term adjustment period. During this period, managers of firms may

148. National Soc'y Professional Eng'rs v. United States, 435 U.S. 679, 691 (1978); Board of Trade v. United States, 246 U.S. 231, 238 (1918); Stratmore v. Good- body, 866 F.2d 189, 194 (6th Cir. 1989), cert. denied, 490 U.S. 1066 (1989). 149. See, e.g., Sicor Ltd. v. Cetus Corp., 51 F.3d 848, 854 (9th Cir. 1995); SCFC ILC, Inc. v. Visa USA, Inc., 36 F.3d 958, 968 (10th Cir. 1994), cert. denied, 115 S. Ct. 2600 (1995); Capital Imaging Assocs. v. Mohawk Valley Medical Assocs., 996 F.2d 537, 547 (2d Cir. 1993), cert. denied, 114 S. Ct. 388 (1993); Schachar v. American Academy of Ophthalmology, 870 F.2d 397, 399 (7th Cir. 1989). 150. See supra notes 94-98 and accompanying text. See also Joseph F. Brodley, Post- Chicago Economics and Workable Legal Policy, 63 ANrrrRusT L.J. 683, 687 (1995)(decrying the requirement of a demonstrated immediate impact on price on the grounds that such a requirement ignores the losses to productive and innovative efficiency caused by anticompetitive practices). 151. BRENNER, supra note 38, at 62-66; Brodley, supra note 150, at 687-88. 152. LEIBENSTnN, supra note 93, at 207-08. 1996] RESTORING RIVALRY TO ANTITRUST take advantage of the shelter of diminished rivalry to allow X-inefficiencies to proliferate, to shirk innovations, or to be unresponsive to customers and fellow employees, rather than to increase price or decrease output. Plaintiffs will not be able to demonstrate an adverse impact on price or output even though diminution of rivalry in the market has had a large impact on producers. A renewed emphasis on rivalry will lead to the abandonment of so- called market power "screens" or "fflters."153 In a Rule of Reason analysis, a court which uses a market power screen holds that a restraint of trade cannot harm competition unless the defendants, individually or jointly, possess a meaningful modicum of market power.' 5 4 The use of market power screens has proven most popular in the area of vertical restraints of trade, although some courts have been willing to use them in cases of horizontal market restraints. 55 The requirement that plaintiffs establish the defendants' collective market power is even less justifiable than the requirement that the plaintiffs demonstrate an effect on price or output. First, the assertion that only firms with market power can reduce output effectively is open to question.'5 6 Second, even if this assertion were true, it wrongly equates competition with output rather than rivalry.'57 Ri- valry can be injured even by firms without market power. If an increased emphasis on rivalry will lead to decreased reliance on or even abandonment of requirements of proof of adverse impact on price, output, or market power, it will lead to an increased sensitivity to the individual business strategies pursued by various firms. Ri- valry is ultimately a psychological phenomenon which is felt in the

153. For information about market power screens and their justification, see generally Frank H. Easterbrook, The Limits of Antitrust, 63 Tax. L. Rav. 1, 19-23 (1984). (Market power analysis should come first, as those firms without market power pose little threat. Those with significant market power may warrant per se analysis.) 154. See, e.g., SCFC ILC, Inc. v. Visa USA, Inc., 36 F.3d 958, 969 (10th Cir. 1994), cert. denied, 115 S. Ct. 2600 (1995); Assam Drug Co. v. Miller Brewing Co., 798 F.2d 311, 315-16 (8th Cir. 1986); Goss v. Memorial Hosp. Sys., 789 F.2d 353, 355 (5th Cir. 1986); Rothery Storage & Van Co. v. Atlas Van Lines Inc., 792 F.2d 210, 220 (D.C. Cir. 1986), cert. denied, 497 U.S. 1033 (1987); General Leaseways, Inc. v. National Truck Leasing Ass'n, 744 F.2d 588, 596 (7th Cir. 1984); Graphic Prods. Distribs. Inc. v. Itek Corp., 717 F.2d 1560, 1568 (11th Cir. 1983); Storer Cable Communications, Inc. v. City of Montgomery, 826 F. Supp. 1338, 1357 (M.D. Ala. 1993), vacated, 866 F. Supp. 1376 (M.D. Ala. 1993)(departure from Rule of Rea- son must be based on economic effect, not formalistic line-drawing). 155. John DeQ. Briggs & Stephan Calkins, Antitrust 1986-87: Power and Access (Part ), 32 AN'TrRUsr BuLL. 275, 285 (1987). 156. Easterbrook's suggestion of a market power filter is criticized in Richard S. Mar- kovits, The Limits to Simplifying Antitrust: A Reply to ProfessorEasterbrook, 63 Ta L. REv. 41, 79-82 (1984) and Robert Pitofsky, Comment: Antitrust in the Next 100 Years, 75 CAL. L. REv. 817, 825-27 (1987). 157. See supra notes 58-59 and accompanying text. NEBRASKA LAW REVIEW [Vol. 75:209 minds of marketplace firm managers. Direct proof of an adverse impact on rivalry would require probing the psyches of the managers of firms-not a very feasible line of inquiry. Fortunately, the amount of rivalry any firm produces in the marketplace can be estimated. The first step in estimating the amount of rivalry a firm generates in the marketplace is realizing that without rivals, there can be no rivalry.1 58 The amount of rivalry any given firm produces in the market is largely determined by two variables, the structure of the market and the business strategy the firm is pursuing. A restraint of trade in a market with few competitors is likely to have a larger dampening effect on rivalry than an identical restraint in a market with numerous effective competitors. 159 This principle is well-recognized in antitrust jurisprudence, including Section One of the Sherman Act.160 Even those who believe that only restrictions on output matter in antitrust laws pay some homage to this principle. Advocates of this position emphasize the need to establish the defendant's possession of market power to prove a violation. Market power is easier to obtain and prove when there is a relatively small number of competitors in the market.161 However, the amount of rivalry generated by a firm is determined by its individual business strategy as much as by the structure of the markets in which it operates. 162 The role of business strategy has, however, been unjustifiably ignored. A firm which pursues strategies which merely imitate those of its competitors generates less rivalrous pressure than a firm which pursues atypical business strategies. A

158. Cf John J. Flynn, Monopolization Under the Sherman Act: The Third Wave and Beyond, 26 ANTrnmusT BuLL. 1, 62 (1981)(without competitors there can be no competition). 159. Cf OSTER, supra note 32, at 211-13 (noting that the larger the number of competitors in an industry and the more similarly the top three or four rivals are sized, the more rivalry and the fewer opportunities for coordination to reduce that rivalry). 160. For example, the starting point for the analysis of the competitive effects of horizontal mergers under Section Seven of the Clayton Act is the effect of the merger on concentration in the relevant markets. United States v. General Dynamics Corp., 415 U.S. 486, 503-05 (1974); Brown Shoe Co. v. United States, 370 U.S. 294, 343 (1962). In a Rule of Reason analysis under Section One of the Sherman Act, courts often examine the competitive effects of a challenged restraint of trade by looking at the effect of the restraint on "market structure including such factors as the number of firms and the market share of leading firms." ANTITusT DEvELoPmENTs, supra note 2, at 48 (footnotes omitted). 161. See supra note 154. 162. Cf OsTER, supra note 32, at 214 (noting that diversity of firms tends to impede coordination which diminishes rivalry); JosEPH L. BOWER, THE Two FACES OF MANAGEMENT 183 (1983)(emphasizing the importance of individual business strategy rather than the number of competitors in determining the competitiveness of a market). 1996] RESTORING RIVALRY TO ANTITRUST 241 firm can choose from among a vast variety of uncommon business strategies. The firm can be an aggressive discounter, or it can appeal to the luxury end of a market. It can adopt an unusual channel of distribution such as mail order sales in a market dominated by con- ventional retail outlets. The firm can make large expenditures in con- ventional research and new product development, or it can explore unusual paths in research and development. In terms of generating rivalry, the details of the strategy are not as important as its differen- tiation from the strategies adopted by competing firms. That differen- tiation is the engine which generates significant rivalry. When a restraint of trade causes a firm to abandon a differentiated business strategy, whether voluntarily or involuntarily, that restraint is likely to do substantial harm to the rivalry which is the essence of competition.163 Antitrust courts should recognize the importance of preserving asymmetrical business strategies for fostering competition. In recent years, antitrust courts have not been sufficiently sensitive to the need to preserve atypical business strategies and have instead fo- cused exclusively on market structure and neoclassical microeconomic theory.164 The Tenth Circuit's decision in SCFC ILC, Inc. v. Visa, USA, Inc. 165 is an example of this focus on market structure and microeconomics theory. This case involved an attempt by Sears Roe- buck, owner of the Discover credit card, to issue a card through the Visa system. The Visa system is a major credit card system made up of individual financial institutions who are the actual issuers of Visa cards to consumers. Sears had obtained control of a bank through which they sought to issue a Visa card. The other member banks of the Visa system blocked Sears from issuing a Visa card by adopting a bylaw of the system which precluded issuers of competing credit cards from joining the Visa system and issuing a Visa card. Sears brought suit against Visa claiming the bylaw was an unreasonable restraint of trade which violated Section One of the Sherman Act. A jury verdict was rendered for Sears, but that verdict was overturned by the Tenth Circuit on appeal.1 66 In reversing the verdict, the Tenth Circuit held that Sears had not presented enough evidence to the jury to allow it to find that the challenged bylaw had, on balance, an anticompetitive effect.167 The court noted that while the Visa venture might have a degree of market

163. cf MICHAEL E. PoRTER, COAnmTrrvE STRATEGY: TECINIQUES FOR ANALYZING IN- DUSTRIES AND ColPzE'rrOas 19 (1980)(noting that competitors with diverse strategies increase the intensity of rivalry). 164. See infra notes 165-204. 165. 36 F.3d 958 (10th Cir. 1994), cert. denied, 115 S. Ct. 2600 (1995). 166. Id. at 960. 167. Id. at 968. 242 NEBRASKA LAW REVIEW [Vol. 75:209 power, all parties stipulated that the relevant market was the system of individual financial institutions which issued the various credit cards under the Visa label.168 No individual financial institution pos- sessed market power.16 9 The Tenth Circuit then proceeded to utilize a market power screen and hold that Sears did not establish that the challenged bylaw had an anticompetitive effect (i.e., increase price or decrease output) because the defendant lacked market power in the relevant market.170 The court bolstered its conclusion by pointing out that a wide variety of rates and terms for Visa cards was available from other banks and financial institutions and that the bylaw did not prohibit Sears from offering a non-Visa credit card, such as its own Discover card, with any rates or terms it chose. 171 The court dismissed as irrelevant evidence presented by Sears that a Visa card issued by Sears would offer consumers a low-cost alternative to existing Visa cards and that the financial institutions voting in favor of the challenged bylaw feared that issuance of a Visa card by Sears would undermine their profits.172 If the court took the concept of competition as rivalry seriously and emphasized the importance of business strategy in producing rivalry, it might have reached a different conclusion. If competition is given its proper meaning as rivalry, market power screens are inappropriate.17 3 The court's alternate grounds for determining that an anticompetitive effect had not been shown are equally inapposite because they show insensitivity to the importance of divergent business strategies. From a purely structural and even personal perspective, the court's holding is readily understandable. Most economists believe that the market for credit cards has all the structural characteristics of a competitive market.174 Moreover, if the judges were like most other mid- dle to upper income Americans, they were probably deluged with offers for Visas and Mastercards offering a wide variety of interest rates and ancillary features.175 On the other hand, if Sears' business strategy is factored into the calculations, then the challenged bylaws

168. Id. at 967. 169. Id. 170. Id. at 969. 171. Id. at 971-72. 172. Id. at 969-70. 173. See supra notes 17-62 and accompanying text. 174. GENERAL ACCOUNTING OFFIcE, REPORT TO CONGRESSIONAL REQUESTORS ON U.S. CREDIT CARD INDUSTRY-CoMP.T1TIVE DEVELOPMENTS 20-22 (1994). 175. Louis Trager, The Credit Game, Dollars and Sense, The Right Deal for Every Kind of Spender? It's in the Cards, SAN FRANcisco EXAMINER, October 31, 1994, at B-1. 1996] RESTORING RIVALRY TO ANTITRUST 243

might well have had a major dampening effect on rivalrous competition in the credit card issuer market. If Sears issued a Visa card, it would likely have incorporated many of the features which were found in its Discover card, i.e., no annual fee, an average annual percentage rate of interest which decreased as purchases with the card increased, and cash rebates for use of the card.176 As the Tenth Circuit implied, some or all of these features may have already been available through issuers of various Visa cards. 177 However, no evidence was presented that this combination of features was available through an existing Visa issuer. Moreover, even if such a combination were available, it was not available in a Visa card aggressively marketed to a wide national audience of credit card users. Sears, in making an unusual and desirable combination of terms and rates available nationwide to a wide base of consumers, would have been pursuing a divergent business strategy. That strategy would probably have greatly increased competition in the form of rivalry in the market for Visa credit cards and would even have affected the price of credit. The increase in rivalry is possible because millions of consumers are apparently not taking advantage of the competitive structure of the market. These consumers spend literally billions of dollars on cards which charge them higher rates of interest or add auxiliary charges such as annual fees, when they could easily qualify for and obtain cards with lower rates of interest and without auxiliary charges. 178 The reaction of existing Visa issuers to Sears' threatened entry is some evidence that this effect would have occurred. Visa issuers were deeply afraid that if Sears were allowed to issue a Visa card, their profit margins would be eroded substantially.17 9 At least one existing issuer also complained that because Sears faced fewer regulatory con- straints, it could undercut the issuer by offering a lower rate of interest on its Visa cards.1so The court of appeals dismissed this evidence on the grounds that the intent of those adopting the bylaw was irrele-

176. Dennis W. Carlton & Alan S. Frankel, The Antitrust Economics of Credit Card Networks, 63 AINrIRUST L.J. 643, 662-63 (1995). The no annual fee provision would be of particular importance to the subgroup of credit card users whom Louis Trager calls "zero balancers--those who carry no interest-generating balances on their credit cards by paying off their balances in full by the end of the back end free ride period. Trager, supra note 175. The interest rebate feature would be of particular interest to the subgroup of zero balancers who Trager terms "Pay as You Go High Rollers"-zero balancers who run up substantial monthly purchases on their credit cards. Id. 177. SCFC ILC, Inc. v. Visa USA, Inc., 36 F.3d 958, 971-72 (10th Cir. 1994), cert. denied, 115 S. Ct. 2600 (1995). 178. Trager, supra note 175. 179. Carlton & Frankel, supra note 176, at 663. 180. SCFC ILC, Inc. v. Visa USA, Inc., 36 F.3d 958, 968-70 (10th Cir. 1994), cert. denied, 115 S. Ct. 2600 (1995). NEBRASKA LAW REVIEW [Vol. 75:209 vant and that businesses generally do not like their competitors or competition itself.18 1 The court missed a traditional and rather obvious point. While the intent of those adopting a restraint is not determinative of its legality, that intent is relevant to assist in predicting the competitive effects of the restraint.18 2 In this case, the reaction of competitors to Sears' pending entry into the Visa card arena indicated that they believed that Sears' entry, coupled with its likely business strategy, would increase rivalry/competition in that market. The area of vertical restraints of trade is particularly rife for a renewed appreciation of the importance of maintaining divergent business strategies in fostering a competitive market. Asymmetrical business strategies such as aggressive price discounting or utilization of alternative channels of distribution are vital to the creation of rivalrous competition. Contemporary antitrust courts frequently have been oblivious to this reality. The Ninth Circuit's decision in O.S.C. Corp. v. Apple Computer Inc.183 demonstrates this weakness. The case involved the termina- tion by Apple Computer of several of its dealers for violating Apple's ban on mail order sales of Apple Computers. O.S.C. and five other dealers brought suit against Apple, claiming that its ban on mail order sales violated Section One of the Sherman Act. The district court granted summary judgment in favor of Apple and the Ninth Circuit upheld the grant of summary judgment.' 8 4 In order to prove a violation of Section One of the Sherman Act, concerted action must be demonstrated.18 5 The court of appeals held that O.S.C. and its fellow plaintiffs had failed to present sufficient evidence to prove that Apple, in terminating the mail order dealers, had acted in concert with any other party.'8 6 Thus, the court held that the trial court did not err in granting summary judgment on the plaintiffs' claim that Apple's actions violated Section One of the Sherman Act.187 This should have ended the court's inquiry. Nonetheless, the court proceeded to discuss why the plaintiffs did not have a claim under Sec- tion One using a Rule of Reason analysis.

181. Id. at 969-70. 182. Board of Trade v. United States, 246 U.S. 231, 238 (1918); LAWREMNCE A. SULU- VAN, HANDBOOK OF THE LAW OF ANTrrRUST 175-79 (1977); ANrITRUST DEVELOP- Emzrrs, supra note 2, at 56. But see Brooke Group Ltd. v. Brown & Williamson Tobacco Corp., 113 S. Ct. 2578, 2597 (1993) (Robinson-Patman Act case rejecting probative value of defendant's subjective estimates of ability to recoup losses sustained in alleged predatory pricing scheme in light of realities of market conditions). 183. 792 F.2d 1464 (9th Cir. 1986). 184. Id. at 1466. 185. Monsanto Co. v. Spray-Rite Service Corp., 465 U.S. 752, 768 (1984). 186. O.S.C. Corp. v. Apple Computer Inc., 792 F.2d 1464, 1468-69 (9th Cir. 1986). 187. Id. at 1466. 1996] RESTORING RIVALRY TO ANTITRUST

The court emphasized Apple's need to eliminate free-riding through its ban on mail order sales.1ss Apple feared that computer buyers would go to full service retail outlets to learn about Apple personal computers, then purchase an Apple computer from a mail order outlet which did not have to bear the cost of familiarizing the customer with Apple computers (or computers in general). The court also noted that the number of retail outlets selling Apple computers increased after the mail order ban went into effect, and the price of Ap- ple computers decreased.189 The result in the case might be correct. Aside from any lack of concerted action, the justification that the restraint strengthened Ap- ple and interbrand competition by eliminating free-riding had particularly strong appeal in this case. The sale of personal computers (especially in the era when the facts of the case arose) requires tremendous point of sales efforts to educate customers and to overcome their fear of technology and computers.' 90 These efforts were subject to free-riding by mail order dealers.191 Moreover, Apple faced stiff interbrand competition from IBM personal computers and their compat- ible clones. Nevertheless, the court might have been too hasty in equating the increase in the number of outlets and decrease in the price of Apple personal computers with an increase in competition in the market for personal computers. O.S.C. and its fellow mail order firms were pursuing at least two divergent business strategies. First, by utilizing mail order sales they an alternative channel of distribution, thereby creat- were providing 92 ing what Joseph Palamountain has termed "intertype competition."' Second, they were pursuing a discount pricing strategy.' 93 The first strategy was, of course, ruled out by Apple's ban on mail order sales. While the second strategy was not precluded by the ban on mail order sales, no evidence was presented which indicated that any conven- tional Apple retailers were engaging in aggressive price discounting. The price of Apple computers did drop after the ban on mail order sales was instituted. The drop in price may not, however, have been a result of enhanced competition among Apple retailers, but a manifes-

188. Id. at 1468. 189. Id. at 1469. 190. David F. Shores, Vertical Price-Fixingand the Contract Conundrum: Beyond Monsanto, 54 FoRDHAm L. REv. 377, 400-01 (1985); Terry Calvani & Andrew G. Berg, Resale PriceMaintenance After Monsanto: A Doctrine Still at War With Itself, 1984 Duz L.J. 1163, 1182 n.69. 191. Mail order dealers did not have to offer extensive customer familiarization and point of sales advertising. Thus, they were able to offer lower prices on the computers, undercutting ordinary dealers by free-riding on the costly point of sales service and familiarization provided to customers. See supra note 190. 192. JOSEPH C. PAz oUNTAnq, Tm POUrTcs OF DISTBUTiON 38-39 (1955). 193. O.S.C. Corp. v. Apple Computer Inc., 792 F.2d 1464, 1468 (9th Cir. 1986). 246 NEBRASKA LAW REVIEW [Vol. 75:209 tation of the general decline in the price of personal computers stem- ming from rapid technological development over the last fifteen years. 19 4 The increased number of Apple dealers is also not necessarily an indication of heightened competition. The increase in the number of dealers would not add to competition if they were all following the same business strategy sheltered from alternative channels of distribution and aggressive price discounting. Indeed, the increased number of retailers might even be a sign that competition has decreased in the market.19 5 The elimination of an alternative channel of distribution and aggressive discounting in a product which was extensively protected by patents and copyrights and which commanded fierce brand loyalty may have eliminated a meaningful amount of rivalrous competition in the market for personal computers. If the court were more sensitive to the importance of divergent business strategies in promoting competition, it might have reached a different conclusion, or, at the very least, relied solely on the lack of concerted action in upholding the grant of summary judgment. While the greatest impact on a renewed emphasis on rivalry will be felt in Section One of the Sherman Act, legal analysis under other provisions of the federal antitrust laws will also be affected. One such area is the analysis of horizontal mergers under Section Seven of the Clayton Act.196 Courts often assess ease of entry into the marketplace when ruling on the legality of horizontal mergers under the Clayton Act. A court which takes rivalry seriously might well not blithely rely on theoretical ease of entry to determine that mergers, including mergers which create monopolies, do not threaten an injury to competition.' 9 7 A merger which leaves few competitors, or worse, leaves only one competitor actually in the market, eliminates most, if not all, salient rivalry. Of course, potential new entrants to the market might supply a sufficient quantity of competition. However, as Michael

194. As each new generation of computers emerges, price drops of ten to twenty percent per year are the norm with substantially larger drops occurring during peri- odic price wars. Wayne Labs, PC Prices: How Low Can They Go, INSTRUMENTATION & CONTROL Sys., Apr. 1993, at 27; Kathy Robello & Stephanie Anderson Forrest, They're Slashing as Fast as They Can, Bus. WEE., Feb. 17, 1992, at 40. 195. See BoPnn, supra note 14, at 45-46 (noting that cartels and price fixing agreements tend to draw more firms into a business). 196. 15 U.S.C. § 18 (1988). A horizontal merger is a merger between competitors "per- forming similar functions in the production or sale of comparable goods or services." Brown Shoe Co. v. United States, 370 U.S. 294, 334 (1962). 197. Some cases which have exhibited this fault are United States v. Baker Hughes, Inc., 908 F.2d 981 (D.C. Cir. 1990); United States v. Syufy Enters., 903 F.2d 659 (9th Cir. 1990); United States v. Waste Management, Inc., 743 F.2d 976 (2d Cir. 1984); and In re Echlin Mfg., 105 F.T.C. 410 (1985). 1996] RESTORING RIVALRY TO ANTITRUST

Porter has found, all rivalry is not fungible. Professor Porter's studies indicate that strong domestic rivalry is more important in fostering world-class competitive industries than foreign competition.19 s The principle which underlies Professor Porter's observation is that rivals who are proximately located generate more competitive pressure than rivals who are more distant. The courts which rely solely on theoretical ease of entry to assume that a merger will not seriously damage competition in a relevant market ignore this lesson. A court which appreciates the importance of strong domestic rivalry would insist on something more than the theoretical absence of entry barriers, e.g., a demonstrated history of effective and successful new entry into the market, to establish that a merger will not injure competition in a relevant market.199 As long as the would-be entrants in a market remain only potential competitors, the psychological pressure they generate, i.e., the rivalrous competition, will not be as strong as that generated by firms that are already in the market. Another example of how paying attention to rivalry would make an important difference in the analysis of horizontal mergers under Sec- tion Seven of the Clayton Act is the issue of acquisition of competitors who are innovators or disrupters in the marketplace. In United States v. Aluminum Co. of America,200 the Supreme Court held that Alcoa's acquisition of Rome Cable, a small competing wire and cable maker, violated Section Seven of the Clayton Act.201 Rome Cable was an innovator and aggressive competitor in the wire and cable market.202 The Court emphasized this point in determining that the merger threatened competition in that particular market. 2o3 The Depart- ment of Justice's 1968 merger guidelines incorporated this point when they stated that a departmental challenge to an acquisition involving a disruptive or innovative competitor in a relevant market "can ordi- narily be anticipated."204 In 1982 the Department's merger guidelines were rewritten. The 1982 merger guidelines demoted the acquisition of a disruptive competitor from an event that would likely trigger a Clayton Act Section

198. PORTER, supra note 109, at 117-18. 199. In Syufy, an existing theater chain did in fact move up into the Las Vegas theater market, reducing the defendant's share of revenues from 93% to 75%. However, whether this was effective entry is another question because the parties stipulated that the other chain was not an "effective" competitor. United States v. Syufy Enters., 903 F.2d 659, 665 n.8 (9th Cir. 1990). 200. 377 U.S. 271 (1964). 201. Id. at 281. 202. Id. 203. Id. 204. United States Department of Justice, Merger Guidelines § 8 (1968), reprinted in 4 Trade Reg. Rep. (CCH) T 13,101 (1995). NEBRASKA LAW REVIEW [Vol. 75:209

Seven challenge to an event which, in close cases, merely made it more likely that the Department would challenge a merger.20 5 Moreover, this would only be a factor in cases where the nature of the market was such that the elimination of a single disruptive competitor would plausibly have an adverse impact on competition. 20 6 The 1984 revi- sion of the 1982 merger guidelines continued the 1982 treatment of the acquisition of a disruptive competitor. 20 7 In 1992 the Department of Justice and the Federal Trade Commission issued new joint Hori- zontal Merger Guidelines.208 These new guidelines devalued the seri- ousness of the acquisition of a disruptive competitor even more. The 1992 guidelines state that the acquisition of a disruptive competitor is merely a factor which makes "coordinated interaction," i.e., collusion, among firms more likely and thus is a potential anticompetitive effect of a merger. 20 9 The various versions of the merger guidelines represent a steady depreciation of the competitive significance of the acquisition of disruptive competitors. This devaluation may be nothing more than a reflection of the infrequency with which enforcement agencies and courts have relied on this factor to attack or condemn mergers. 210 The infrequent reliance on the acquisition of disruptive competitors to block or unravel mergers may, in turn, be caused by a lack of such acquisitions in the real world211 If this is true, then the apparent increase in tolerance for the acquisition of disruptive competitors may be little more than symbolic. Nonetheless, the watering down of the perceived anticompetitive grav- ity of acquisitions of disruptive competitors is unfortunate because it sends the wrong message to firms in the marketplace. Market disrupters provide rivalrous competition greatly disproportionate to their size or number. These market disrupters provide the diversity of business strategy which constantly pressures the managers of other firms and refreshes competition in the market. Their disappearance, through means other than failure on the merits of their product or service, usually results in a serious diminution of rivalry in the market. Courts interpreting Section Seven of the Clayton Act should rec-

205. United States Department of Justice, Merger Guidelines § 111C2 (June 4, 1982), reprinted in 4 Trade Reg. Rep. (CCH) 1 13,102 (1995). 206. Id. 207. United States Department of Justice, Merger Guidelines § 3.44(c) (June 14, 1984), reprinted in 4 Trade Reg. Rep. (CCH) S 13,103 (1995). 208. United States Department of Justice & Federal Trade Commission, Horizontal Merger Guidelines § 2.11 (April 2, 1992), reprintedin 4 Trade Reg. Rep. (CCH) T 13,104 (1995). 209. Id. 210. Jay Greenfield, Beyond Herfendahl: Non-Structural Elements of Merger Analy- sis, 53 ANTITRUST L.J. 229, 243 (1984). 211. Id. Jay Greenfield speculates that maverick market disrupters may be more unwilling to be acquired than other firms. Id. 1996] RESTORING RIVALRY TO ANTITRUST

ognize this reality and give little leeway to acquisitions of disruptive competitors, even if such acquisitions are rare and appear to be sys- temically unimportant. While a restoration of the promotion of rivalry as a central concept in antitrust law will result in a number of major changes in contemporary antitrust doctrine and the outcome of cases brought under the antitrust laws, that restoration cannot and should not be grounded on a monomaniacal devotion to rivalry. To carry through on this Article's earlier analogy to a restored monarchy, rivalry must remedy its past excesses and compromise with the new political realities which arose during its period of exile. The changes which should and must be made are the subject of the next section of this Article.

V. REFORM AND COMPROMISE A. Reforms-Differentiating Between Injury to Competitors (Rivals) and Injury to Competition (Rivalry)- Adoption of a Concept of De Minimis Injury to Rivalry If rivalry is to remain a viable organizing concept in antitrust law, a distinction must be drawn between injury to rivalry and injury to rivals. This idea should not be a shocking proposition to even the most zealous proponents of rivalry. After all, it was the populist-oriented Warren Court which, summarizing the equally populist legislative history of the Cellar-Kefauver Act, first stated that the antitrust laws protect "competition, not competitors."212 The distinction is important to draw for two reasons. First, if no distinction is drawn, the line between state unfair trade practice law and federal antitrust law will be eradicated. A successful unfair trade practice directed against a competitor, by definition, injures that competitor. If such an injury is automatically equated with an injury to competition, then every effective unfair trade practice injures the competition which is supposed to be the central concern of the federal antitrust laws. Counsel for competitors injured by unfair trade practices could bring a federal antitrust action on the basis that "competition" has been injured by the challenged act. The lure of bringing a federal antitrust action is the availability of treble damages and attorney's fees under the federal antitrust laws. This would result in the federal courts facing a flood of cases currently brought in state court. As a practical matter, the availability of federal antitrust remedies would render state unfair trade practice claims and remedies superfluous and subsume most of that body of law under the umbrella of the federal antitrust laws. As a number of courts have pointed out, this re-

212. Brown Shoe Co. v. United States, 370 U.S. 294, 320 (1962). 250 NEBRASKA LAW REVIEW [Vol. 75:209 suit is neither desirable nor within the intent of Congress in passing the federal antitrust laws.2 13 Second, the distinction is important because rivals can be injured without any injury to rivalrous competition. Indeed, if one rival injures or even destroys another through competition on the merits of its product or service, rivalrous competition as a process has been validated, not injured. The adoption of a concept of de minimis injury to rivalry would seem to be a separate "reform," but in fact, it is inextricably linked to separating injuries to rivalry from injuries to rivals.214 A distinction between injury to rivals and injury to rivalry is not workable without a concept of de minimis injury to rivalry. Any injury to a rival injures rivalry to some extent. Assume, for example, that two firms are vying for a contract. One firm conspires with a legally independent party to spread disparaging falsehoods about its competitor. The falsehoods are believed by the firm awarding the contract and the contract goes to the rival who disseminated the false information. No other contracts are affected and the firms remain bitter rivals for other business. What the winning firm did certainly constituted a tort.215 Because of the disparagement, the managers of the victorious firm probably felt less psychological pressure to be responsive to the needs of the firm awarding the contract. In some sense rivalry is diminished. If competition is equated with rivalry, a violation of Section One of the Sher-

213. See, e.g., Military Servs. Realty, Inc. v. Realty Consultants of Virginia, 823 F.2d 829, 832 n.4 (4th Cir. 1987); Sutliff, Inc. v. Donovan Cos., 727 F.2d 648, 655 (7th Cir. 1984); Northwest Power Prods., Inc. v. Omark Indus., 576 F.2d 83, 90 (5th Cir. 1978), cert. denied, 439 U.S. 1116 (1979); Merkle Press, Inc. v. Merkle, 519 F. Supp. 50, 54 (D. Md. 1981). 214. The use of the term "reform" in the text is not meant to imply that modem day antitrust courts provide remedies for insubstantial injuries to competition. In- deed, many antitrust courts (including the Supreme Court) have already articulated the view that the antitrust laws are not concerned with trivial, insignificant, or de minimis injuries to competition. E.g., United States v. Topco Assocs., 405 U.S. 596, 606 (1972); Smith v. Pro Football, Inc., 593 F.2d 1173, 1183 (D.C. Cir. 1978); Doctors Steuer & Latham v. National Medical Enters., 672 F. Supp. 1489, 1504 (D.S.C. 1987); United States v. National Ass'n of Broadcasters, 536 F. Supp. 149, 158 (D.D.C. 1982). The term "reform" is merely meant to reject the extreme, albeit logical, implication of restoring rivalry to a central place in antitrust law-any diminution of rivalry violates the antitrust laws. Such an approach would rule out, for example, developments such as the merger of small rivals in a market overflowing with vigorously competing firms or the sharing of technological know-how among competitors because these actions, to some small degree, diminish the rivalry between the parties to the arrangement. Applying federal antitrust law in such instances would not only be overkill, but would, in the long run, hinder both economic efficiency and rivalry. 215. For information on the tort of commercial disparagement, see generally 2 FOWLER V. HARPER ET AL., THE LAw OF TORTs §§ 6.1-6.4 (2d ed. 1986); W. PAGE KEETON ur AL., PROSSER AND KEETON ON THE LAw OF ToRTs 962-77 (5th ed. 1984); 2 RUDOLPH CALLmANN, TE LAw OF UNFAIR COMPETITION, TRADEMARKS AND MONOPOLIES §§ 11.01-11.35 (4th ed. 1994). 1996] RESTORING RIVALRY TO ANTITRUST

man Act may have occurred. Legally independent parties have restrained transactions (created a restraint of trade) which has injured "competition." Moreover, the restraint has no conceivable procompetitive or other justifications. 216 Nonetheless, this hardly seems like the case for the imposition of federal antitrust liability because any successful business tort is apt to diminish rivalry at least to the extent described in the above example. Yet, the only way to escape the "logic" of the conclusion that the antitrust laws have been violated is to require that the diminution of rivalry reach a threshold of significance before it constitutes a violation of the antitrust laws. The precise parameters of a significant injury to rivalry are not easy to establish. In addition, tremendous variations may exist among courts in judging whether an act or practice meaningfully injures competition. These difficulties are worth confronting because the alternatives are unacceptable. One alternative is to simply abandon rivalry as the defining concept for competition. This is both bad law and bad economic policy. Moreover, substituting output or any form of economic efficiency for rivalry will engender similar problems. Net output is often difficult to measure and sometimes even difficult to deflne.217 Even if these concepts can be measured, some minimal threshold injury to output/economic efficiency will still be needed to avoid burdening the courts with trivial injuries that are really not the concern of the antitrust laws. Another alternative is to simply state that competition is injured any time rivalry is diminished in any degree. Under this approach, automatically allowing an antitrust suit for any injury to rivalry could punish actors for pursuing rivalry and competing on the merits of their products or services. Given the consequences of these alternatives, the task of ascertaining the boundaries of de minimis injuries to competition is an onerous, but necessary task. B. The Grand Compromise-Giving Economic Efficiency a Co-Equal Role Drawing a distinction between injuries to rivals and injuries to rivalry and establishing minimum thresholds of diminution of rivalry are needed to make rivalry a viable central concept in antitrust law.

216. The spreading of false information is a deadweight social loss. RicHAPM A. Pos- NER, Tm EcoNinc ANALYSIS OF LAW 109-10 (4th ed. 1992); E. Thomas Sullivan, On Nonprice Competition: An Economic and Marketing Analysis, 45 U. Prrr. L. REv. 771, 791 n.101 (1984). For an argument on how such tactics can, but not necessarily do, injure competition, see Harry S. Gerla, FederalAntitrust Law and the Flow of Consumer Information, 42 SYRACUSE L. R!v. 1029, 1052-80 (1991). 217. Cf Thomas G. Krattenmaker & Steven C. Salop, Anticompetitive Exclusion: Raising Rivals' Costs to Achieve Power Over Price, 96 YALE L.J. 209, 283-84 (1986) (discussing the ambiguities and limitations of measuring the concept of output). NEBRASKA LAW REVIEW [Vol. 75:209

In this sense, no inconsistency exists between these modifications and rivalry. Conceding that gains in various forms of economic efficiency can offset losses in rivalry is a very different matter. By definition, allowing gains in economic efficiency to offset losses in rivalry places rivalry in a subordinate role. Proponents of the centrality of rivalry can argue that such a concession is unjustified as a matter of statutory interpretation. 218 They can also point out that recognizing a direct role for efficiency considerations is a form of "double counting." Eco- nomic efficiencies and other desirable social goods will be created if rivalry is promoted. Finally, if economic efficiency can justify a diminution of rivalry, why shouldn't other values, such as the maintenance of countervailing power, the preservation of small independent businesses, the spreading of economic opportunities to minorities, or the diffusion of educational opportunities, be permitted to offset losses in efficiency? All of these points are analytically sound. The only response is that concern with the various forms of economic efficiency is so en- trenched in the courts, and perhaps in the mind of the public, that an antitrust policy which does not directly address this concern would lack the appearance of legitimacy it needs to be sustainable in the long term. In recent years, both the Supreme Court and the lower federal courts have continually emphasized the need to consider economic efficiency in gauging whether practices pass muster under the federal antitrust laws.219 A great deal of contemporary dissatisfaction with traditionalist/populist antitrust doctrine stems from a belief that the doctrine has hindered the ability of the United States to compete in a global economy it no longer dominates.220 The beliefs that rivalry im- pedes efficiency and that an antitrust policy based on the promotion of rivalry hinders the global competitiveness of American industries are perversely wrong.22 ' Nonetheless, faith in the need for promoting economic efficiencies is so strong that an antitrust policy which does not recognize a direct role for claims of efficiencies probably will lack the judicial and public support needed to make it effective. Why should efficiency be promoted and not other values? The sim- ple reason is that while other values have occasionally enjoyed judicial recognition as being able to offset competition,222 they do not enjoy the

218. See supra notes 7-60 and accompanying text. 219. See supra notes 4, 45. 220. Phillip Areeda, Antitrust Law as Industrial Policy: Should Judges and Juries Make It?, in ANwrRusT, INNOVATION AND COMPETITIVENESS 29, 34-35 (Thomas M. Jorde & David Teece eds., 1992). 221. See supra Part III. 222. See supra note 46. 1996] RESTORING RIVALRY TO ANTITRUST

same widespread and persistent allegiance as efficiency. This answer may not be principled, but it is pragmatic. What then should be the role of economic efficiency in an antitrust regime centered around rivalry? Efficiency should be granted equivalent status with rivalry when it comes to weighing procompetitive effects. In other words, gains in the various forms of economic efficiency should be considered benefits to competition. This stance may not be justifiable as a matter of logic or statutory interpretation, but it is practical politics. Those who urge that courts once again accept impact on rivalry as a vital consideration in the evaluation of an act or practice under the antitrust laws already face a Herculean task. Convincing courts to abandon their commitment to consider directly economic efficiency under the antitrust laws is an impossible task. While accepting a direct role for claims of economic efficiencies may be a political necessity, such acceptance need not be uncritical. Specu- lative or theoretical claims of economic efficiencies should not be permitted to counterbalance demonstrated diminutions of rivalry.223 Unfortunately, the record of courts in taking a critical look at claims of economic efficiencies has been far from encouraging. For example, in the area of vertical non-price restraints of trade, courts have often un- thinkingly accepted claims that the restraints will eliminate free-riding without enquiring whether a free-riding problem actually exists or whether the challenged restraint will solve the problem.224 In the area of horizontal mergers, courts, and sometimes even enforcement agencies, have too often relied on theoretical ease of entry into a market to offset the anticompetitive effects of a merger, rather than insist- ing upon actual entry by competitors into the market.2 25 The spotty record of courts in critically evaluating claims of economic efficiencies ought not stand in the way of a compromise which recognizes such efficiencies as coequals of rivalry. If a court is unwilling to evaluate the alleged economic efficiencies with a skeptical eye and an open mind, it will probably be unwilling to accept the importance of rivalry in the antitrust laws. The grand compromise will do no harm because furthering rivalry is already a lost cause in those courts.

223. The approach used by the Department of Justice in its 1982 and 1984 merger guidelines relating to efficiencies claims can serve as a model. The guidelines adopt a skeptical attitude and demand clear and convincing proof that efficiencies can offset the anticompetitive effects of a merger. Department of Justice, supra note 205, § VA & n.53; Department of Justice, supra note 207, § 3.5. 224. See James F. Rill, Non-Price Vertical Restraints Since Sylvania: Market Condi- tions and Dual Distribution,52 AqrrRusT L.J. 95, 101 (1983). 225. See supra notes 197-99 and accompanying text. NEBRASKA LAW REVIEW [Vol. 75:209

VI. CONCLUSION The decline in the primacy of rivalry as an antitrust value has been continuing for almost a quarter of a century. Reversing that decline will not be an easy task nor will rivalry ever assume the unique position it enjoyed prior to the 1970s. Changes and painful compromises will have to be made. Nevertheless, the task is worth undertaking and the changes and compromises are worth making because the stakes are enormous. Rivalry is vital to encourage managers to remove wasteful inefficiencies which cost producers and consumers billions of dollars each year.2 26 Perhaps even more important, the maintenance of strong domestic rivalry is crucial to the development of industries which can compete globally.227 Trillions of dollars, millions of jobs, and the standard of living of the nation ride on the development of such industries. As Professor Porter has observed, vigorous enforcement of appropriate antitrust laws is the most important policy governments can pursue to promote that rivalry.2 28 The size of these stakes should encourage those who wish to see rivalry restored to undertake the difficult task of convincing courts to take rivalry seriously, and to make the sometimes distasteful changes and compromises necessary to restore the promotion of rivalry to its rightful place in antitrust law.

226. See supra notes 91-93 and accompanying text. 227. See supra notes 125-29 and accompanying text. 228. PORTER, supra note 109, at 662-64.